The AM Forum

I finished electrical work on my Valiant yesterday and did initial power up tests today – absolutely dead.

Prior to starting restoration I verified the transformers were good etc.    I solid-stated all 3 supplies.   I used socketed diodes / dropping resistors for the bias and LV supplies.   Those units, as well as all other tubes, were out of the radio when I powered up today using a variac and low amperage fuse.  My AC current meter showed no current draw at all, and no bulbs came on.    J8 had the plug in.

I replaced the internal plate supply fuse holder with one on the chassis,  and of course the plug mounted fuses with one on the chassis.   I was very careful doing this work,  but obviously I got something wrong.   

Here's a diagram of what I currently have:

I'm really wondering if the wire from T2 to RY1-2 should go instead to the Filament ON/OFF switch;  I'm pretty sure it didn't before I started, but now I'm now sure.

The two wires labeled “unknown” go the wiring harness near the modulators and I've been as yet unable to trace them.

Any help appreciated.

First, you need a good schematic...

Grab one here: http://www.mbzponton.org/n2awa/radio_Johnson_Viking_Valiant_schematic.jpg (http://www.mbzponton.org/n2awa/radio_Johnson_Viking_Valiant_schematic.jpg)

Second, your schematic shows that T2, the low voltage power supply transformer, will never see 117vac. There are several discrepancies between what you have drawn and the schematic, so it would be good to do a comparison and be sure that things are wired as they should be.

Assuming that the power supply hasn't been rewired and you still have the original plug, check that the two fuses in the actual power plug are good. Johnson fused both leads of the A.C. plug with internal fuses.

Did the Valiant work fine before you began the modifications?

Yes, I have a schematic (excellent manual & schematic from Pete ("manualman") .

I have no idea if the radio worked.   As I said,  before starting work I verified transformer secondary AC voltages with all tubes removed.    To that extent,  the radio at least powered up and the Filament ON/OFF and Plate ON/OFF switches worked as expected.

I've stared at that schematic for a number of hours (the power input circuit isn't that hard.... ::) ).

I agree that T2 will never see 117 VAC so I had to have rewired something incorrectly.    Removing the wire that now goes to RY1-2 to the hot 117 should solve the problem.   However,  I don't think it was there when I started my work,  and I'm concerned that moving it may induce other issues.   

I was hoping someone had a construction manual that shows what's connected to those three coils and can confirm moving that wire is the correct solution.

Check this out; it wont load for me all the way

http://hammadeparts.jivetones.com/Amateur_Radio_Manuals_Schematics/JOHNSON/Johnson%20Viking%20Valiant%20Transmitter%20(construction%20instructions)%20WW.pdf


I ended up using Firefox to view the file

klc

Thank you;   that's exactly what I was looking for!

Good luck with the restoration and all the challenges you are likely to encounter...be patient, this rig will test you but it is rewarding to get them going.

Found the problem -- after finding it, I thought of the saying “if it was a snake it would have bitten me!”.

I did the preliminary VFO tests as per the construction manual.

80M (VFO Full CCW):
OSC                      22 mA
Buffer (1/3)      10 mA
Grid (Exciter)       6 mA peak

40M (VFO Full CCW)
OSC                      25 mA
Buff (1/3)              25 mA
Grid (Exciter)       7 mA peak

With my solid state rectifiers and 120VAC input:
Bias is -250 VDC   (pin 5 of SS plug in)
LV  is +372 VDC  (pin 8 of SS plug in)

Note I have not yet adjusted the bias pots as per the manual.    

Haven’t turned the HV on yet either…. Tired now, so best not to play with HV…..

Getting excited about maybe having the Valiant on for the AM Rally!

Congratulations on the progress!

You could use a variac or series 150W bulb to limit the current draw if you're concerned about the first HV test. In any case, start with the Drive level low for first test to prevent slamming current while the final tuning and loading are adjusted. If the plate current pins with the drive down then the Clamper Tube adjustment will need to be made first.

Looking forward to hearing it on the air!

Greetings from central PA where we're having a heat wave -- 25F !!

The HV turned on with mostly no ill effects.  I switched to AM and the two VR tubes lit up.

As I was staring at the schematics with the radio on (HV off) I heard a small "pop".   I removed power but could find no zorched components etc.   I applied power and did the same VFO testing I did last night on 80 and 40M,  and they were fine, the bias, LV, and HV looked fine.   I'll have another look tomorrow.

I performed the RF Bias adjust as per the manual.

Since I have not yet tuned up the radio,  I left the AM Bias pot mid range.

I adjusted the clamp level as per the manual.

I was rather concerned about the LV and bias values as I built solid state rectifers with dropping resistors.

Here's some measurements:

V11-6 (manual):  330V
      (120 VAC): 358V
      (115 VAC): 326V

V11-9 (manual): 5.8 VAC
      (125 VAC): 6.2 VAC
      (115 VAC): 5.6 VAC

HV    (Manual):  600V
      (125 VAC): 716V
      (115 VAC): 652V

Since I didn't use a dropping resistor for the HV solid state rectifers, I figure that's good for about 15 to 20V more....

Bias Rectifier (R50 side not connected to SS rectifer):
      (125 VAC):  -216V
      (115 VAC):  -194V

After doing the clamping adjustment,  I find when I turn the HV on the Ip spikes to around 400 mA.  When I turn the HV off the Ip spikes to aroud 200 mA.  Is this normal on the Valiant?

I think it's time to squirt some RF tomorrow!

Welcome to the "El Valiente Club". You'll get used to this kind of thing after a while...


AM bias?

Having the RF bias properly set to -70 Volts prior to tuning up is important. I use the non-grid end of L7 as a test point.

The modulator bias is also important to be sure that those 6146 tubes aren't in full conduction. I adjusted the bias for about 60mA of modulator resting current.


Not that I've experienced. The 6AQ5 is supposed to conduct enough to sufficiently pull down the screen supply to the 6146 RF tubes whenever RF excitation isn't present. The clamper adjustment can be tricky and may need some fine tuning -or- this could be due to insufficient bias voltage.


That would be fantastic!

There be RF here!  :D :D :D

I turned on the Valiant and after warmup checked the oscillator current -- nothing.  On any band.   Nuts.   Poked around and couldn't find anything.    I then remembered that I had plugged in my straight key into J2.   I wondered if J2 was a shorting type jack, so hit the key and the Oscillator current was fine.   Whew.....

The good news is I'm getting about 200W out on 80 and 40M.   The bad news the Ip is pegged and will not dip.

After playing around with this a bit I put the Meter into the Grid position and proceeded to adjust the "Final" control and the power did indeed peak.   Adjusting the Coupling controls got me to about 210W out on 80M and about 190W out on 40M.    I'm wondering if the problem is the meter shunt?  I replaced both the original plate and modulator current shunts so they should be good.   

Here's my front panel readings:

80M:
Osc:  25 mA
Buffer:  18 mA
Grid 8 mA    (6.5 max key down)

40M:
Osc:  25 mA
Buffer:  15 mA
Grid 7.5 mA    (6 max key down)

Since everything looks to be in the ballpark, and the power peaks with the Pi Network controls, I really don't think there's a problem with the Ip not dipping - I think it's the meter shunt.  Another thing that leads me to this conclusion is that I put the dummy load into J5 initially -- I was seeing some dipping around 300 mA -- which I think is wrong for no RF load.   Not seeing any power on the wattmeter and 300 mA on the Ip was pretty much a turn the radio off now situation and take a look. As I said, I did replace both meter shunts and they tested OK.

I tried to adjust the Modulator resting current (of course I could not load to 330 mA so I went for max power out).   I could only adjust down to 75 mA, not 50 mA as per the manual, and that's at the extreme end of R61.   For the RF Bias adjustment, I added a 1K (I think...) resistor in series with the R62 wiper; I read that this helped in the adjustment process due after solid stating.    I may do this for the R61 wiper arm as well.

I'm also a bit concerned that I can't get 8 mA drive during key down conditions.

Comments welcome from you Valiant gurus.....

It's pretty important to get the plate current meter to read accurately, or, get the dip where it should be. Have you tried 20M? Failures of the output loading caps will prevent low enough current dips at lower frequencies. Also, turning the drive down should reduce the plate current. Does it?

Do you have access to an antenna analyzer? There's a way to use that to check the output tuning network.

As for the modulator current, that should easily be able to be lowered if the bias supply is solid stated since it's producing too high a voltage. Not being able to get the static current down to the normal level suggests something else is going on.

Did the current spike issue get resolved?

Good morning.    The same for 20M.    Yes, turning the drive down reduces the Ip;   to get 300 mA of Ip the drive does not register on the meter and power output, on both 80 and 20M, is around 25W or so.


Yes, I have an MFJ-259 analyzer.   No, I don't know how to use it to check.


Can't disagree with you;  the common item for both currents though is the new meter shunts.   Maybe I'm fixating on those to much?


No, still there.

Output tuning networks are basically impedance matching circuits and they work both ways so it's easy and safe to troubleshoot component issues using the following method.

First, the transmitter should be unplugged and all power supply caps checked for full discharge (that's the safe part).

Next, a non-inductive resistor with a value that matches the RF section is connected between the plate leads and chassis. For the Valiant, a 1.8k or 2.2k resistor will work fine. 1/4 Watt is more than sufficient. Leads should be as short as practical.

An antenna analyzer or impedance bridge is then connected to the antenna jack and set for the frequency to test. The most rigorous test would be on 160M as that utilizes all of the tuning components.

The Final, Aux, and Fine controls are then adjusted to get a 50Z reading with +/- 0 reactance. The settings should match what you use for that frequency during normal operation. If you can't get a match or if the controls are at their maximum ranges then it's likely that a loading cap has failed, which is common, particularly for the turnstile cap.

Cool, eh?

The RF shunt is R58 (0.202 Ohms) and the modulator shunt is R59 (0.404 Ohms), but since the modulator current sample also flows through R58 then an issue there could cause high readings for both MOD and Plate. Check that the R58 resistor wire connections are tight. To be absolutely sure, you could lift one end and use a current meter of known accuracy in its place. For calculations, I believe the Valiant's meter is 5mA full scale.

Hi Clark

Sorry for the delay in response;  everyone in the family is catching the bug that's going around and I'm no exception.    I went downstairs today and ran tests on 40M and that was about it for me.

I placed a 2.2K resistor from the final plate cap (junction) to ground and then tuned up via the Analyzer.   Here's what I got:

R= 51.5 ohms      X = 10.0 ohms

Final:  78
Coupling Aux: 0    Fine: 7

Max power output was 130W with 7 mA drive.   

Ip still way off the scale, but would come in range with decreased drive.

I then tuned up via the manual, removed power etc, then put in the 2.2K resistor and hooked up the analyzer:

R= 90 ohms   X = 38.8 ohms

Final:  90
Coupling Aux: 3  Fine: 0

Max power output was 200W with 2.5 mA drive.    I did start out with the Coupling and Fine controls both at 0.

Increasing the drive in both situations did not yield increased power output.

Both cases show a lack of C to cancel out the L,  but is 10 ohms that critical?    Not sure what's causing the 38.8 in the second case.

The resistor value should be the same as the impedance of the RF tubes. Your B+ is significantly higher due to the solid state rectifiers and that nominal line voltage is higher today than it was in 1957 and the actual current is unknown without a working Plate current meter. That said, I can understand the discrepancy.

A more significant issue is the high amount of loading capacitance needed for 40M. You might find it difficult to tune on 80 and 160M. This suggests a loading cap failure.

I think it's pretty important to get the current meter working since the Valiant wasn't designed to operate at these voltages. You might see rated output, but that could mean that energy is getting dissipated in the components.

Check the lug connections for the meter shunt resistors, particularly R58 (0.202 Ohms). Soldering nichrome wire isn't an option so the lugs are clamped onto the ends of the resistance wire. Sometimes the lugs become loose. Be sure the chassis is clean at the contact point.

Hi Clark

Thanks for you comments!

Both the plate and modulator current shunts were replaced during the restoration.  The 0.404 ohm modulator shunt was replaced with two 0.82 ohm resistors in parallel and the 0.202 plate ohm shunt was replaced with four 0.82 ohm resistors in parallel.  Not the best solution but I used what I had on hand.

I tested the meter during restoration.   I measured 16 ohms internal resistance and 5 mA full scale deflection.    I asked Rodger NQ9E to confirm the resistance as 16 ohms was a bit odd, and he said the nominal resistance is 20 ohms.   So, I think the meter itself is working OK.   I put a pair of 1N4004 diodes across it for protection,  although that's not near the best solution with 0.7 volts forward drop and the meter dropping 100 mA full scale, but again it's what I have on hand.

To show 500 mA on the meter,  the shunt will need to drop 495 mA.

The plate tranny T1 output is 720V and the input is about 120.  So  Ns/Np = Vs/Vp.
Ns/Np = 720/120 = 6.

Ip/Is =Ns/Np  Ip = Is(Ns/Np)

For 300 mA plate current,  Ip = 0.33(6) = 1.98A which is within specs of the tranny. (and Ip in this context is primary current)

But, the meter is being pegged rather hard.  If the meter is seeing even 7.5 mA though it,  the total current would need to be about 740 mA.

Primary T1 current = 0.740*6 = 4.4A and that would most certainly blow the 2A fuse in the plate primary circuit.   While I've blown one or two,  if the current was that high I'd be blowing it every time I tuned up.

So, my thoughts are the plate current really is in range but is being displayed incorrectly -- yet I've replaced the shunts.   Maybe I'm fixating on this?

Maybe I'll disconnect the T1 CT where it connects to C92 and stick a meter in there.....

It's better to put a pair of diodes in series, then back to back across a meter.  Otherwise, as the diodes come into conduction, you end up with bad values being shown on your meter. You don't want any current flowing through the diodes until a fault.

This somewhat hinders the diodes protecting the meter, but.....

W6WRT (Bill Turner) showed this effect on the amps reflector years ago.

--Shane
KD6VXI

That would seem to get things close as long as the resistors have close tolerances so at this point you've got the arduous task of troubleshooting numerous components, ie, meter, shunts, power supply (assuming the off-scale readings are correct...yikes).

Did the original shunts have to be replaced? That's not something to do unless it's absolutely necessary. It may be worthwhile to put the original shunts back in unless they are known to be bad.


Kind of...

I tested the meter on my Valiant and determined that it showed 0.1 Volts across the terminals at full deflection. In series with a ammeter of known accuracy, it showed 5 mA, confirming that the internal resistance is 20 Ohms.

*** These tests were done with the meter in the "Off" position to remove any circuitry.

In theory, the shunt will deliver 0.06 Volts across the meter at 300mA of plate current. If you're seeing a meter resistance of 16 Ohms then that voltage would produce a reading of 375 mA on your meter.

Yep. Get out your magnifying glass and begin searching for clues to lead to the suspect!

hear hear for those good manuals!

I was told to replace the meter shunts as they were usually not as accurate as modern components, so I did.   They were thrown out long ago.

I've a question on "SW6", the Output Coupling switch, on the schematics.  What does the "rear" section do?   There's only 2 capacitors,  but nothing is connected to them!! 

SW6, the "Aux Coupling" switch, connects various combinations of 300pF, 600pF, 900pF, and 1200pF in the "turnstile" capacitor to the output circuit for loading. The variable capacitor shown (C9) is the "Fine Coupling" adjustment. You can test this by putting an Ohmmeter between the output lead and each of the turnstile cap tabs and observing the continuity as the "Aux Coupling" control is rotated. On 160M, all the caps should be connected.

Send pix!

Could the front and rear rotor sections be connected integrally? Could there be one wire connecting the two sections' common contacts? One of those would make sense for adding capacitance in steps.

I should have clarified my post;  I know what SW6 does,  I just don't see how the rear wafer does anything as the two caps C42B and C42C are not connected to anything.

There has to be some common contact between wafers.   I'll take a look.

61F When I left the office and we have a winter storm watch for tomorrow morning -- ice and snow mix.... :P :(

You think the two caps are not connected to anything.  Take a close look at the rotating wipers.  Most likely the rear one is connected to the front rotating wiper by way of the taps that hold the wipers in place on the rotating insulating material.

Fred

Mark....the first thing that caught my eye was over 210 watts output.
With the Ip at 400ma and 700v on the plate that's 280 watts input.
If that is really true....why is a fuse not blowing?
If there is no Ip dip...what frequency is the output on?
Sompin weird here.

Gents

Thanks for your replies, both on this site and personal.     Saturday morning I rooted through the "pile-o-meters" and surprisingly found I have a 0-1 A DC meter.    I performed some tests and verified it was working.     Other things going on Saturday, including (unfortunately) me getting sicker.   Flu, cough etc.

I didn't think it the best idea to go mucking with the HV until I felt better.   So it may be a day or so before I disconnect the HV and verify the plate current reading.

Dave -- you're comments about why the 2 Amp fuse is not blowing is one of the reasons I'm wondering if it's the meter or shunt,  please see my previous post about this.   

Good Evening.

Tonight I inserted a 1 A meter at the output end of L43.

I was able to dip the Ip as expected using the external meter, but the Valiant meter was pegged on Ip;  it was fine on the other settings.

Some tests into a dummy load;  this time I started with the recommended settings for Aux and Fine as per the manual chart.

160M:
180W Output @ 450 mA Ip  and 8 mA ig
Aux: 7     Fine 3
Exc 55     Final 78

80M:   Blew a lot of plate fuses…..   with only a single 2A fuse left  I moved on….. ;D

40M:
190W  Output @ 400 mA Ip and 4 mA Ig  (ig Maxed out)
Aux 6     Fine 0
Exc 48     Final 60

I guess I’ll need to remove the shunts and inspect them.    I may just order 0.2 and 0.4 ohms resistors anyway.

I will play with 80M and the other bands tomorrow or Saturday.

I’m thinking of just replacing all the caps in the Pi Network.   I haven’t priced that out yet.

73 Mark K3MSB

Any chance of placing your test meter in just the RF plate circuit? One possibility is to lift the jumper between pins 2 and 6 on P8, the 9 pin accessory plug on the back, which would give you a much more accurate reading. As it is now, you're seeing the total current of the AF and RF plates, screens, bleeder resistors, voltage regulators, and the surge current from the HV supply caps each time you key up. Yikes, actually.

Weird that 80M is blowing fuses and 160M is not. I say that because the latter uses all of the loading caps, so the low bands are usually the first to not dip. One thing to consider is that the oscillator runs at 160M with the 5763 (V4) driver acting as a doubler for 80M operation. Are you able to get 7-8mA of grid current on 80M?

 

I put the meter where I did as I wanted to get a total picture of what was going on with the plate current.   I thought the plate component would far outweigh the others for what I was looking for.

I was thinking of putting the meter at the junction of L12 and R16,  but putting it in the jumper as you suggest looks to be the same.

One thing I've noticed is that with an initial drive of 8 mA you need to be fast on dipping the plate;   any particular reason the manual starts out at 8 mA instead of 2 or 4 mA for the initial dip?

I will revisit 80M tonight or tomorrow -- when I've restocked 2A fuses.

 

Ah-HAH! 7.5mA is the target, not the starting, grid current...

The best way to tune up is to begin with the Drive control set somewhat low, then bring it up so that you get reasonable plate current, such as 300mA, then start dipping. The Drive control is no longer advanced when it gets to where it should be, verified by clicking the meter to Grid current. Use the Aux and Fine loading controls to further increase the loading until the dip is at the right current then double-check the grid.

BTW, at zero drive, the clamper will/should limit plate current which will gradually increase with drive.

Yep, sounds like a hassle, but this is what separates the "appliance operators" from those with "The Right Stuff" as Chuck Yeager would say.

Well, not according to the page 14 of the manual!

     Turn METER switch to "grid", tune EXCITER dial for maximum grid current.
     Adjust grid current to 8 mA by turning DRIVE control.
     Turn METER to "plate"

     Immediately after throwing SW8 to "man" position (on) tune final dial for minimum current (resonance).

After blowing enough fuses,  I modified that to an initial grid current of about 2 mA or so then upped it afterwards.

Makes me curious what the E.F. Johnson engineers considered "immediately". I would have used the term "At the speed of light".

My Valiant didn't work right because the original builder followed the assembly instructions to the letter.  In fact, I spent -four months- puzzling over why I had anemic grid current on 20M only to find out that a later revision of the manual included a modified driver tank circuit.

Love these rigs, but sometimes....

I have absolutely no experience, rpt no experience with the  Valiant, but that won't stop me from commenting....

The Johnson Viking du,  Manuel has one do all the Osc and buffer peaking and then dropping the drive ( grid current on da meter) to zero, before switching on the HV. Then throw the hv switch and watching the plate meter, advance drive to around 100 mA, then dip. Then recheck the grid current. Maybe a nicer way of loading the thing up.


klc

Well, unexpected results.

With the 1A meter between pins 2 and 6,  the difference between that reading and the one taken before (after L43) is about 1 needle's width.    That's seems too small. ???

That *is* odd given how many other current drawing devices are no longer being measured.. At least your meter isn't getting slammed when you key up. On the other hand, a 1A meter isn't going to indicate much difference if you lose the 60mA of modulator resting current.

Tried a 500 mA meter.     Results a bit more realistic.

Meter between J6-8:
400 mA.
50 mA Resting current
 
Meter at the output end of L43:
430 mA
80 mA Resting current.

That seems more reasonable.

I'll be hopefully working on the Valiant on and off this weekend;  stay tuned.....

Getting output on all bands and didn't blow any fuses today, but I'm concerned about some readings on the higher bands.

Band  OSC(ma)    Grid(ma)   Ip(ma)   Pout(W)   Exc  Final  Aux  Fine

160       25               8          450       180         55    78      7       3
80        25             1.5-2        430       200         58    70     3        7
40        25                4           400       190         48    60     6        0
20        27              .5-1         420       125         51    62     9        2
15        27                1           420       160         25    32    10       5
10        30               2            420       180         42    22   10        3

The grid current is maxed out on 80, 40, and 20M               
               
The grid current can be increased a bit on 15M and 10M but no further output               
               
On 10M, VFO set to 28.0,  my frequency counter showed output on 27.5.   Verified on Icom RX;  VFO needs aligned.
               
The plate current is measured with external 500 mA meter between J8 pins 2 and 6.   

The drive on 15 and 10 seems very low for the power output;  is this typical for a Valiant?            

That's a milestone for Valiant owners. You should be picking out a bottle of bubbly...

The grid current readings suggest a weak 5763 driver/multiplier tube and you should be able to get 8mA on all bands. The Drive control varies the 5763 screen grid voltage, so if you get to a point where no further advancement of the control increases grid current then the tube is weak.

What are the "Buffer" readings for each band? That measurement is the cathode current of the 5763 and can indicate tube condition. As the manual says, the Buffer current should be around 22mA when the Drive control is at "3", which is read as two times the bottom meter scale reading (11mA).

Just checked the buffer current on 10M since the radio was already tuned up there from this afternoon -- about 16 mA with the drive on 3;  my measurements on 10 were made with the drive on 5 and it was 22 mA.

Do you have a small coil on your bandswitch where the wires connect from the driver tank? It might be time to revisit a post I made just about a year ago (reprinted with my own permission)...


Suffering from low drive?

Several Valiants have come across the workbench with this issue, so I decided to share the secret to restoring that Valiant libido to where it should be and with only minor invasive surgery!

"Valiant A", as we'll call it, had nearly no drive on 20M. "Valiant B" had widely varying drive on the upper bands. In both cases, the 5763 would not be able to create the specified 7.5 mA of PA grid current, and high settings of the DRIVE pot would cause saturation.

As it turns out, later factory wired Valiants and all Valiant IIs have an additional coil "L6C" mounted to the SW3A band selector switch which increases the inductance and Q of the driver tank circuit.

L6C can be easily made with 12 turns of insulated solid conductor hookup wire tightly wound on a 3/8" diameter (I used a drill bit). Installation is simple:

1) Red wire from driver tank coil removed from terminal #4 on SW3A. Insulate the end as it will not be used.
2) Green wire from driver tank coil moved from terminal #5 to terminal #4 on SW3A.
3) L6C installed between terminals #4 and #5 on SW3A.

The attached schematic shows the change of heart the E.F. Johnson engineers had late in the original Valiant production run that became the norm for the Valiant IIs.

You should see an improvement in drive for 20M and up or your money back. Not recommended for 11 meter operation. Use as directed.

Hi Clark

Thanks for the mod for the drive issue;  that seems straightforward to implement.

Before looking into the shunt issue (which will require some surgery) I decided to try the Valiant on AM just to make sure there were no big issues lurking there.   

I did not run at the full 330 mA Ip as I'm still leery of the plate current issue,  but the AM sounded OK on the ICOM with the the audio up to the recommended modulator current.   The clipper worked, but I don't think I desire to use that too much.....

OK, so I need to get the shunt issue solved, which is what I'll start on tomorrow.

It would be great to hear it on the air and we're a reasonable distance apart. Drop me a note if you like to set up a sched. 40M is good in the early afternoons and 75M is good until about 5:30pm.

I load my Valiant to 300mA for AM. Plenty of power and much easier on the modulator, particularly with today's higher voltages.

Good Evening All

I thought I had the plate current problem identified, but alas that was not the case.

I found a cold solder joint at the ground end of shunt resistor.   My meter read about 0.8 to 1 ohm from before the solder lug joint to ground, and 0 afterwards.   

The screw it's attached to secures a terminal strip, and the user didn't even use a lock or star washer.   The top is blocked by a transformer so the screw can't be removed as I wanted to put a toothed solder lug next to the chassis.    I decided instead to attach a solder lug to the screw nearby securing the transformer itself;  it comes with a nice star washer and I added a regular washer.    My meter now reads a solid 0.2 ohms from the hot end of the shunt to ground.     Interestingly, when I was re-attaching the nut to secure that terminal strip,  the bolt sheared off.    Sometimes you just can't win.

Anyway,  the Ip is still high.    Using my external meter connected across the plug on the back,  I'm finding the Johnson meter reads about 175 mA high.     I don't know if that was true before or not.

I also cleaned the meter switch contacts.

I'm getting kind of stumped.   There's not much else to try.   If the meter resistance (which I measured a while back at 16 ohms) is indeed accurate,  then the 0.2 ohm shunt will no longer work, nor will the modulator shunt.    I was thinking of didling with the shunt to try to bring the meter back in line using my external meter as a reference.

If the meter is not reading correctly then that's where the correction should be whether it means obtaining a new meter or adding a resistor in series with the meter to calibrate it. What happens if you put your known good meter in series with the Valiant meter? They should read the same current with 5mA full scale. You could also try putting a meter in place of the Valiant meter.

Good Evening

The beating of the head against the wall continues. :P

I decided to put a 0.9 ohm resistor in parallel with the 0.2 ohm Ip shunt.   This dropped the Johnson current meter reading by about 30 mA.      I then put a 0.2 ohm resistor in parallel with the 0.2 ohm Ip shunt and this dropped the Johnson current reading by 50 mA.    Putting in a lower parallel resistance did nothing more;  I even tried using resistance wire with the length such that it was almost negligible.  There must be something else in the circuit that has a low enough resistance that is taking over as the lowest resistance;  not sure what.    In any event,  using a lower value shunt does not look to be the answer. 

I then read Clark's post.

I set the Ip to full scale and then inserted my 500 mA external meter in series with the Johnson meter.   The Johnson meter dropped to less than half scale.   Hmmmm.     Well, if the internal resistances of the two meters was about the same that might be expected.

I then put my Fluke DMM in series with the Johnson meter.    The Johnson meter read full scale and the Fluke read 4.99 mA.   Great!    I then used the Fluke to measure the voltage across the Johnson meter and it was about 103 mA.   Great.    That shows an internal meter resistance of about 20 ohms, as it should be.

But then it started getting “not so great”......

I decided to verify my readings and put the Fluke back in series with the Johnson meter.   Now the Johnson meter will NOT go full scale.   Removed the Fluke and the meter goes full scale.   I repeated this with my Simpson 260 as well as my Military ME-29U Ammeter.   

I've no idea why I got good readings once then not again.   

Here's some photos of the HV rectifier board and the LV and Bias solid state plug-ins.

If you put your Fluke meter in series with the Valiant meter, do they agree on other meter switch settings? It's possible that the Valiant meter windings or movement is damaged. Isolating the meter readings issue to the meter or sample points will determine the next step. You're looking to see if they either basically agree on the various switch settings.

Studying the schematic, the shunt for the modulators is 0.404ohms.  The shunt for the PAs is 0.202ohms.  You can not insert another meter just anywhere to put in series with PA current metering circuit.  Doing this will disturb the metering circuit and you will get false readings.

There is one location in the circuit where you can do this.  The connection point where the PA current shunt connects to the plate xfmr CT.  Keeping the meter circuit connection on the 0.202 shunt, open the connection between that connection point and the plate xfmr CT.  You can put another current meter between those two points without disturbing the metering circuit.  The inserted test current meter will read the same current that the plate current circuit is reading.  You can compare current readings to see how far off the plate current metering circuit is.

Depending on how all these components are located and wire in the circuit, doing the above may not be that easy.  You may have to remove connections from one connection point and separate the one circuit point into two separate connection points.

Study the schematic that is posted on the first page of this thread and you will see what I am talking about.

Fred

Good Evening Clark

Yes,  I looked at that and the meter seems to read fine:

             Fluke      Johnson
Grid        1 mA         5
Buffer     1 mA         5
OSC       1.7 mA       8.5

I'm using just the bottom scale in this case for all 3 settings.
5 corresponds to 1 mA
25 corresponds to 5 mA etc

"5" corresponds to 1 mA.
5/25 = 0.2
0.2 * 5 mA full scale = 1 mA

"8.5  corresponds to 1.7 mA  
8.5/25 = 0.34
0.35* 5 mA full scale = 1.7 mA

Hi Fred (KA2DZT)

I think we're talking about two different things.    I first put my external meter between the output end of L43 and the junction of C91/R17.   This reading was a bit high and someone pointed out that this would include the bleeder current etc.   Clark then suggested I move the external meter to be between J8-3 and J8-5 so as to get rid of the bleeder component.    This meter consistently reads the same and seems to be correct.     I think this is equivalent to what you suggest.

Clark posted "What happens if you put your known good meter in series with the Valiant meter? They should read the same current with 5mA full scale"    I misinterpreted this to mean the Johnson meter reading would not change if I had it set to 450 mA (full scale) before inserting the meter.    Thinking about this,  although the new meter (or DVM etc) is now in series with the Johnson meter,  BOTH are across the 0.202 ohm shunt.   So, the lower readings I though were incorrect were probably correct;    I need to ensure that whatever the Johnson meter is reading matches what the inserted device is reading --  which will not be the same as the standalone Johnson meter.

So, I have one data point that shows 100 mA across the meter full scale with my DVM measuring 5 mA which is correct;  how the devil I configured that I'm not sure -- I have a lot of wires hanging out there.....
       

Another thing to try is to set your Fluke to volts and measure the voltage drop across the Valiant meter. The benefit is that it won't affect the meter reading. Once you have a cross-reference between it's current reading and your voltage reading, you'll be able to see if anything changes like it's done during the last test.

With no DVM’s attached,  I set the Ip to 500 mA (FS) and noted the external meter indicated 335 mA.

I put the DVM #1 in series with the Johnson meter, and DVM #2 across the meter.

This means DVM #1 and the Johnson meter are both across the 0.2 ohm shunt.

DVM #1 indicated 3.28 mA
DVM #2 indicated 64.4 mV.
Johnson Meter showed 335 mA (Ip scale).

Johnson meter is 335/500 mA = 0.67% deflection.

Current should be 0.67 * 5mA = 3.25 mA – this is what  DVM #1 shows.

Voltage should be 0.67 * 100 mA = 67 mA – this is what DVM #2 shows.

Granted the current and voltage is wrong as I have a DVM and the Johnson meter across the 0.2 ohm shunt,    but they are correct for a 335 mA Johnson meter indication.

The external Meter shows about 335 mA Ip, which in this case does not correlate to the 335 mA Johnson meter reading. The external meter is hooked between J8-3 and J8-5

I’m starting to conclude the meter itself is OK.   Comments?

So the question is if the meter readings are accurate for all the meter switch settings. If so, then you just need to find a stable resistance to put in parallel with the shunts to make the meter readings accurate for the parameters that are off.

From one of my previous posts:

I decided to put a 0.9 ohm resistor in parallel with the 0.2 ohm Ip shunt.   This dropped the Johnson current meter reading by about 30 mA.      I then put a 0.2 ohm resistor in parallel with the 0.2 ohm Ip shunt and this dropped the Johnson current reading by 50 mA.    Putting in a lower parallel resistance did nothing more;  I even tried using resistance wire with the length such that it was almost negligible.  There must be something else in the circuit that has a low enough resistance that is taking over as the lowest resistance;  not sure what.    In any event,  using a lower value shunt does not look to be the answer.

That last part bothers me.

I've removed the modulators so the 0.404 ohm shunt shouldn't enter into the issue.

There's nothing else in that circuit.    Maybe remove C52 off the Clamp tube?   But if that was leaky,  it in parallel with the 0.202 shunt would cause the effective shunt to be smaller and the meter reading to be less, not more.

 ??? ??? ???

At this point, I say forget the Valiant. We need to find some dingy dive-bar with a dirt cheap beer special, scorching hot wings and a huge plate of nachos and exchange Valiant war stories. Actually, I've heard all of yours from the beginning so I'll do most of the talking, lol. You can pick up the first round...

Time for those other Valiant geniuses to chime in..

I would recommend that you recheck the ground connection on the 0.202 shunt.  The slightest amount of resistance on the ground connection will cause the PA plate current to read high.  That's why when you keep reducing the shunt resistor with parallel shunts it stops making any reduction in the reading....

Have to take another look at the schematic,  I'll be back

Ok, I looked over the metering switch.  Again, make sure the PA current shunt is grounded.  To eliminate any possible grounding problems,  run a ground wire from the PA shunt ground lug to the ground lug for the ground on the meter switch.

Make sure there are no cold solder connections.  Resolder the ground connections with fresh solder.

Once you have a 100% good connection between the PA shunt resistor itself and the metering switch you should get correct readings.  Slightly poor connections on the meter switch will cause the meter to read low not high.  But, the only thing that will cause the PA current to read high is a poor ground connection at the PA shunt or a shunt resistor that has increased in value.

I also looked at the octal plug connections.  Placing a test current meter between J8-3 and J8-5 should read the same PA plate current that is passing through the shunt.  That meter will be in the B+ voltage line so you have to be careful.

Fred

Good Morning Fred


The ground connection has already been completely redone.


I can try that.   I also thought about just moving the entire shunt atop the chassis near the meter switch.


As said, the ground connection has been rebuilt.   I'm pretty sure I checked for cold solder connections elsewhere,  but reflowing them won't hurt.


This is where I have my external meter connected.

While staring at the schematics I noted the HV should be 600V.    Mine is between 650 and 716 depending upon line voltage.  However, with the divider composed of the two 20K resistors and the 0.202 shunt,  that voltage swing shouldn't have any impact across the shunt.

All basic food groups!

Only takes 0.05 ohms of contact resistance between the metal chassis and the ground lug and the 0.202ohm shunt is 25% too high.  This will cause the PA plate current to read high.  Soldering a ground wire from the shunt ground lug (even solder to the wire on the resistor) to the ground lug at the switch will eliminate this problem.

Fred

Are we collectively sure that the Valiant plate current meter reading is correct and can be confirmed with an external meter in place of the jumper on the accessory jack (intercepting the plate circuit)?

If that's the case then it's really just a matter of determining why the current is high. Test the 6146 screen voltage. R16 is a large 12k resistor that could have drifted in value.

Don't think it's reading correctly.  He said, in a recent post, that the meter is reading 500 ma while the test meter is only reading about 335 ma.

Fred

Agreed. And that could also be within the margin of error for the resistance measurement of the shunt resistor or the resistor itself. Running a separate ground wire is logical, but it would be better as a test than as a solution.

I have a digital meter and I doubt that its accuracy is that good at low resistances simply because it uses such low voltages for testing. It's very easy to get an erroneous reading unless you're dumping a bit of current through the test component so that the voltage drop can be more accurately measured.

Yesterday's test had a DVM across the Viking meter (which is the same as across the shunt) and it showed 64.4mV while the Viking meter read 335mA. So, R=E/I (0.0644/0.335=0.192 Ohms) The caveat is that you also had another DVM measuring current in series with the Viking meter. Could you try that again with just the Viking meter showing current and a DVM reading voltage across the meter?

Theoretically you should see 60.6mV across the Viking meter when it reads 300mA (0.202*0.300). I'm dismissing the Viking meter's 20 Ohms for simplicity.

Yes, and I mentioned that in one of the above posts.   BOTH the DVM and Johnson meter were across the shunt,  so the Johnson meter reading would not agree with the external meter across J8.

Also, regardless of whether the Johnson meter read as expected,  the voltage across it was always what was expected based upon it's deflection.

Tonight I will run the new ground wire.   I'm not sure where on the chassis the meter switch is grounded,  but I know I didn't check that particular ground.


I tried this for 500 mV,  but I can certainly try for 300mV.

This morning I placed an order for a 0.2 ohm shunt resistor;  not 0.202 but that's the closest I could get.  Insanity momentarily gripped me as I considered paying $29 for next day delivery;  but then I realized I wouldn't have any money left for beer, nachos, and wings, so I elected a cheaper shipping method.   

You need a Wheatstone Bridge to accurately measure the 0.202ohm resistor.  I have 4 of them.  You don't need to waste time playing with all the DVMs.

Probably didn't need to order more shunt resistors.  Anything you order may not be exactly 0.2ohms anyway.

Does the meter read correctly on the other ranges??  Is it only the PA plate current that reads high??

If it is only the PA plate current that is reading high,  the problem is the ground like I told you.

Fred

You tellem Freddy, just keep that FMing under control.....

W3GL... ;D

Hi Ralph,

Fixed the FMing years ago.

Fred

Tracked the issue down  :) to a bad solder joint where the shunt connected to the terminal strip.    Which I checked before..... ::)

Johnson               External
Meter (mA)          Meter (mA)
100                      150
200                      230
300                      310
400                      390
440                      420

These were on 160M.

I reflowed the solder connections on the meter switch etc and that did not significantly alter any readings.

I will exercise the other bands tomorrow.

Nice! The 300mA range is the most important and that's where the meters seem to agree. I wouldn't strive for the book values of current, particularly since your voltages are so high. I use 300mA for AM and 330mA CW. No sense in stressing out a 60 y/o transmitter.

Was that cold solder connection on the ground side or the high side of the shunt??

Anyway glad you found it.

I based my comments assuming the solder joints on the shunt were properly soldered leaving only the ground to the chassis as the problem.

Fred

The bad joint was on the high side of the shunt.   My 0.202 ohm shunt are 4 resistors in parallel;  a 0.2 ohm resistor is en-route from Mouser so I'll revisit the shunt next week.

I'm not sure the VFO is reading correctly so I may need to futz with that a bit as well.

Next week I'll do the driver mod Clark suggested.   There's also a mod for the VR tubes so that the Modulator current reads correctly;  I'll see about doing that next week too.  Audio mod will come afterwards.

I plan on being on for the AM Rally this weekend,  so today and tomorrow will be spent getting a 2nd station up and running with the Valiant being the transmitter.   I like 160M DXing,  but I've never been on 160M AM so we'll see how that goes.

Clark and Fred,  thanks so much for your help..... and keep reading this thread.... no telling what lurks in the future!
 

Moved the Valiant to the operating position in anticipation of trying it out in the AM Rally.

Turn on the HV and blowing fuse F1 to the LV transformer T2 like candy.    Put it back on the bench,  same thing.     Did absolutely nothing to it.

Grid drive is set to 0 so I'm not overdriving it;  in fact, I never go key down to tune up!  

Tried it on 160, 80, and 40.   Same thing.     I removed J8 and that had no effect.  

 ??? ??? >:( >:(

What size fuses are you using??

Still blows fuses with J8 out??

Let me know

Fred

Using the recommended 2A fuses, yes with J8 out.    I'm real glad I moved that 2A fuse to a chassis mount fuse holder!

I was really frustrated and just a tad PO'ed, so I just put the radio back on the bench and called it a night.   I take a look at it on Saturday.

I noticed that the VFO was slipping so I was playing with it, removed the dial face etc.   Readjusted the pointer, reinstalled the dial face and tried to tune up.   Started blowing fuses as soon as I turn on the MAN/PTT switch;  don't even need to key down.

With J8 out I'll start by isolating the smoothing choke and see what that shows.

I looked at the schematic.  I see three fuses.  Two 8amp fuses on the main AC line and a 1.8amp fuse on the primary of the main power xfmr.  Exactly what fuse are you blowing??  I'm not seeing a plate fuse but I'll look again.

One thing you should know,  those 866 HV rectifiers are a common cause of shorts.  If you had the xmtr on its side or upside down, those 866s will short as soon as you put voltage on them.  Mercury rectifiers have to be kept in an upright position at all times.

The original power cord has been replaced with a three wire grounded power cord, with a single fuse in the hot side (8A) -- that would correspond to F3 on the schematics. 

F1 is the 2A fuse, thought it's shown as 1.8A on the schematics.   I have no idea where to get 1.8A fuses around here.

The high voltage rectifers have been solid-stated.

This sounds like classic J8 shorts. Specifically, either a disk bypass cap has shorted or one of the pesky TVI coils whose insulation has rubbed through. Kevin, WZ8J, another Valiant restoration jockey, experienced a bypass cap failure at J8, and this reminds me of what he went through.

1) Unplug and discharge.

2) Put an Ohmmeter between the + lead of one of the filter caps and chassis.

3) Start wiggling components near J8. If that doesn't do it, unsolder a lead somewhere to start isolating the HV circuitry.

You're earning your doctorate degree in Valiantology!

Most likely a 2amp fuse is a little light for that xmtr.  OK FB on the 3 wire cord and the single 8amp fuse in the hot leg, that's all correct.

The 2amp fuse should be a slow blow type.  If it is not a slow blow type, that could be the reason they are blowing.  When you hit the PTT, the relay puts power to the HV xfmr.  If that is when you blow the fuse,  check the all those stupid and unnecessary .005 bypass caps on the AC line. check the plug jack for the antenna relay.

Also make sure there isn't a wiring mistake where the 2amp fuse is in series with HV xfmr primary instead of just the LV xfmr primary.

Looked at the schematic again.  You mentioned that you moved the 2amp fuse holder to the top of the chassis.  Make sure that when you did that, that you didn't make a wiring mistake and you have the current for the HV xfmr also passing through the 2amp fuse.

Fred

Are those bypass caps and coils really necessary?   I've had the urge to just rip them all out;  It's not like I'm going to interfere with Channel 2 anymore.....

I get what fuses I can;  slow blow fuses aren't stocked too much around here.   If it's not a Home-Depot, Lowes, or an auto-parts store,  that's pretty much it around here.   I'm pretty sure the fuses I exhausted last night were slow-blow.    At any rate, I haven't replaced one all through the shunt resistor "ordeal",  so whatever I had was working.

I will start tracing tonight.

I found a VFO issue.   Turning the VFO does not necessarily change the frequency;  I may have to turn it an eigth of a turn for it to actually change frequency;  that seems to depend where on the dial it is.    I looked into the VFO cage and the cap is turning and moves throughout it's range.   Not sure what's going on with that.    I was pretty careful about the coupler,  never resting the radio on it's front panel.     I was looking into that when the fuse blowing started.......

Those TVI coils and caps would have been the first thing I would have ripped out.  They are just a possible cause of problems.

The 2amp fuse is only suppose to be on the LV xfmr primary.  Should have nothing to do with the HV xfmr or the HV circuits.

I've built many power supplies (I have more than 1000 xfmrs and chokes) over the past 55+ years.  Without measuring the primary current I think a 2amp fuse is a little too light for that xmtr.  I would use at least a 3amp slow blow, ,maybe even a 4amp fuse.  There is a 8amp fuse on the main AC line.

A little rough figuring,  700VDC x 500ma = 350 watts for the HV supply.  Add 20% for xfmr loses. that's about 420 watts.  420 watts/120vac = about 3.5 amps primary current for the HV xfmr.  

The schematic shows less than a 2amp fuse for the LV xfmr.  If you add up all the loads on the LV xfmr and do the same rough figuring you will probably find that the primary current on the LV xfmr primary is at or near 2 amps.  You would need a fuse greater than 2amps.

Another thing,  with today's higher AC line voltages the primary current on these old xfmrs can increase do to core saturation increasing the xfmr loses.

Another thing,  if you replaced the PS filter caps with caps that have more capacitance. that will increase the start-up in-rush current causing fuses to blow.

Fred

Fred --

I posted incorrectly (which I just corrected in a previous)......    I posted

"Turn on the HV and blowing plate fuses like candy"

That should have been

Turn on the HV and blowing fuse F1 to the LV transformer T2 like candy.

I will check my wiring in that area tonight.   I may "harvest" those coils and caps too......

Interesting if I have a wiring error in that section and it's worked so far without incident.   Will let you know.

Well this is a new one...

So, activating the HV supply is overloading the LV supply. Here are my guesses...

1) There's a short causing the HV to feed into the LV supply - Check the 6146 plate voltage when the LV is on but the HV is off. Could occur at J8 where the supply wiring converges.

2) If the diodes that replaced the 866 high voltage regulators are wired to plugs, perhaps they are wired to pin 4 instead of pin 1 on the 5 pin plug. This would put the heater wiring (derived from the LV power transformer) in series with the HV supply. Turning it on would induce a pulse of current back into the LV transformer as the HV caps charge. While the heater winding is only 2.5V, a solid surge of current would be transformer up in voltage, causing momentary saturation.

Okay, okay, long shot, but it had to me mentioned. Why it's happening now and not before kinda defuses (pardon the pun) the theory.


The AM Rally starts in 5 hours. No pressure, though, just sayin'.

I was checking the filament wiring on the schematic.  All the audio tube filaments are returned to "Y" on the LV xfmr.  All the RF and other tube filaments are returned to "H" but I can not find the feed for "H" anywhere on the schematic.  Is there another filament xfmr somewhere??

Have to take another look maybe I missed something.

Fred

Not seeing anything for the "H" filaments

"H" connects to terminal 8 of J8 via a small coil and the circuit is completed to the power transformer from there. This allows the RF heaters to be disabled if the unit is used solely as an audio amplifier.

Dupe

What Clark said.

See J8.

Phil - AC0OB

Ok now I see it,  looked at everything except the jumper on the octal plug and I looked close at the socket just not the plug.

Well,  the old girl is up and running and putting out RF!!

There were three issues.    First,  when I moved the LV fuse holder F1 to a rear chassis mount fuse holder,  I did not wire up the new fuse holder correctly.   Second,  I incorrectly labeled the fuse holders on the inside when I was soldering the connections.

After fixing that, I turned on the HV and no fuses blew;  repeated a few times.

I then removed power from the unit and remounted the HV bulb I3 assembly.   Applied power and blew the 8A fuse the second time the HV switch was thrown.    To make a long story short (pun intended  ;D)  one of the lugs from the bulb mount was shorting to ground.

So,  will continue to test tonight and see.

I need to get to the VFO Shaft Coupler inside the VFO housing through a hole in the chassis bottom.....

I'm still going to try and get the Valiant in the AM Rally for at least one QSO.....

By the way, at least one end of the filament lines to the LV and Bias rectifiers are physically disconnected.   For the HV circuit, the filament lines are as they normally are,  but I do not use those tube sockets pins with the SS replacements.

News as it develops.....

Good making progress,  I suspected something wasn't wired right with those fuse holders.

Good thing you're not working on my HB xmtr.  It has about 15 power supplies and I think about 30 fuses along with about 35 relays.  Don't ask it's a long story.  Built in stages, it took about 8-10 years to design and build.

Fred

I found out how to tighten the VFO coupler.   It was challenging, but once I found the two setscrew through that tiny hole,  it wasn't too bad.

If I put the VFO at either extreme and rotate to the other extreme there's no slipping as verified by watching the tuning capacitor.   If I stop anywhere in between,  more than likely there's slipping – several inches on the dial -- especially if I change direction.

When the dial is slipping I can visually see the setscrews rotating,  but the tuning cap is not turning.

I tried loosening and re-tightening the coupler screws and that has no effect.   I was going to loosen them, rotate the shaft (or tuning cap), then re-tighten.    That shouldn't cause any harm (as I need to realign the dial pointer anyway), but thought I'd ask first.

Maybe the tuning cap needs a little WD-40.

Mark,

Thank You  ( and the other contributors on this thread ) for detailing your Valiant rebuild .

Your details will be valuable when I start working on my Valiant.

I purchased a good looking Valiant at a hamfest , but I haven't had the time to even open the case for a visual inspection.   

I'm guessing that it still has the *low-fidelity* interstage audio transformer that could be replaced with a tube phase splitter if necessary.

I see mention ( Timtron ? ) that the 6146s don't make good modulators and that triodes like  807s would be a better match for the three RF 6146s, but I haven't found any schematics for this modification.

Anyway............ Thanks again to everyone who is helping to keep the Vals
active on the bands.

Don W4DNR

   

1st QSO with the Valiant !!!

W1TS on 80M in the AM Rally!!

The modulator current meter is really sensitive to the audio gain, and I'm not thrilled with what the scope is showing -- not getting near 100% modulation.   

73 Mark K3MSB

Sounded great! Nice to hear the rig on the air (finally, lol) but seriously, the effort has paid off. Keep the plate current modest on AM and you'll have no trouble reaching full mod and use the clipper to keep the current peaks in check.

Looking forward to the next contact!

Greetings!

I’ve been busy with non-radio stuff this week but I look forward to getting back to the Valiant tomorrow.

I had a great time in the AM Rally even though I was only able to get on 75 M.    I was going to go to 160 Saturday evening but realized the SX-101A didn’t have 160 capability (duh...) and I was too lazy to haul the R-390A over.   The Valiant’s VFO is way off on 40M – full CW deflection is 7280 – so 75M it was.

I participated in the AWA Net on Sunday afternoon and ran into Jim WA2MER that I haven’t talked to in ages.   I found the comments from the net members to be rather diverse and I wish I had time to hang around for “Round 2” to comment on the comments, but was unable to do so.   

Back to Mark’s Valiant saga.

The issue with the dial slipping was the rear screws on flexible shaft coupler.  The front screws can be reached fairly easily, but I could not reach those rear screws.  After pondering options, I drilled a second one-half inch hole just behind the existing chassis hole, slightly offset so as to be under the tuning shaft.  I then snipped the “bridge” between these holes to make a larger oval opening.   It is now very easy to access the fore and aft set screws and the new opening is big enough so you can shine some light inside while working.    Drilling holes goes against my mantra of not doing irreversible mods, but the hole is inside the VFO cage and will not be seen.   

So the radio is basically squirting RF and the plate current seems well behaved now that I redid the solder connections for the meter shunt.    Things to be done:

Dial tracking / VFO alignment on 40 and higher is off.    I have the alignment procedure for that.    Dial tracking / VFO alignment on 80 and 160 is almost spot on.

The modulator resting current is about 25 mA too high with the adjustment pot at its limit.   I plan to add a dropping resistor in series with the adjustment pot.    I added a dropping resistor in series with the RF adjustment pot and that worked nicely.

Add the new inductor to increase drive as Clark N1BCG suggested.

Perform the mods to the two VR tubes so the modulator current reads correctly.

Add a new plate current shunt resistor – the “real” one came in from Mouser this week…..

Will keep y’all posted!

What's the bias voltage on the output of the supply? The manual calls for -265 but you should see a bit more. I had an issue like this and it turned out to be C93 having gone bad.

I'm leery of adding components to fix issues. There's some problem getting masked...

Not a lot of radio time this weekend.

What time I did have I built a signal "pick off box" -- which may not be the correct name.    Basically an aluminum box with 3 SO-239 connectors -- input, output, and "pick-off".   The pick-off is coupled to the input/output wire by being about 1/4 inch from the other wire.    Very simple, and I think it was K1JJ that had the original idea.    I have it going into my frequency counter as I was told a few times during the AM Rally that I was a bit off frequency.

Last night I got the VFO working correctly on 40 and higher.  I need to squirt RF to be sure all's OK.

Clark,  OK on your comments.   I will recheck my voltages after the SS replacements.

All the electrolytics are new,  and the HV ones are bit more stout.

If you want to monitor your frequency, the pick off point needs to be pre-modulation. Sideband energy from sampling the RF output will confuse your counter.

You could use the SSB adapter input jack, an SO-239 connector on the rear of the chassis, if you don't plan to use the Valiant as a linear amplifier. Just unsolder the connection that's there, and run shielded wire to the 6146 RF tube area.

The connection would be to the grid circuit through a D.C. blocking cap and a resistor divider (there's way more PtoP voltage there than you need). A 0.001uF cap would work. The divider values would depend on the voltage your counter needs but it's a good idea to use a high value resistor in series with the cap and a low (50-100 Ohm) resistor for the shunt at the input of the shielded wire.

Yep.   I have to use an unmodulated carrier or else the frequency counter doesn't stabilize.   I wanted a solution that is transmitter non-specific.   Also, I do a lot more CW than I do AM.

Is the -265V bias voltage on the schematics correct (possibly from model run to model run) ?

With a 170 ohm dropping resistor,  I measure -225V at the downstream end of R50,  and -261 volts right at the end of the diode stack.   Current is 32 mA.

With no dropping resistor, I measure -241V  at the downstream end of R50,  and -272V right at the end of the diode stack.   Current is 34 mA.

I did not replace R50, so it had to be in tolerance.  C93 and C98 are new.   Line Voltage is 115V.    Measurement made key up,  HV off.

I removed my voltage dropping resistor from the solid state replacement for the 6BY5G.

This caused the reading at the downstream side of R50 to go from -225V to -240V.    If anyone’s interested, the current is 34 mA measured directly after the diodes.     I’m not sure where the -265V value on the manual schematics comes from; I’m running the radio at 115V, so maybe it was at a higher line voltage?

I am now able to adjust the modulator resting current to be within spec; 65 mA with the control at the extreme of one side.    This is still concerning, but at least it’s in spec.

Drive on 80M is still 5.5 to 6 mA.    I tried another 5763 and got the same results.     I do not have the shunt resistor R56 marked as replaced as out of spec, but I will recheck that.

Someone asked about the LV.   I measured this at J8-4 with the transmitter tuned up on 75M AM

B+ OFF:        354V Key up,  314V Key Down
B+ ON:         345 Key up,    298V  Key Down.  

I will look at making the L6C coil next.

I built and installed L6C as suggested by Clark.

The old and new grid drives are as follows:

Band   Old Grid(ma)   New Grid (ma)  Pwr Out  Drive control
160        8              8                     125           4
80        1.5-2          4                     150           9 (Saturated)
40         4              4                     160           9 (Saturated)
20        0.5-1          6                    160            9 (Saturated)
15         1              4.5                  140            9 (Saturated)
10         2              4                     150            9 (Saturated)

Ip was 400 ma for these tests.   Got dips on all bands.

I not quite sure what's up with 160M;  The aux control was at 10 and the Fine control at 5;  moving them around did very little.  This might be cockpit error on my part.

The manual says 8 mA drive on all bands;  is this really what's to be expected?   Perhaps an increase in L6C?   The drive being saturated isn't good.

The modification to remove the regulator current from the Ip Meter setting was already done to the radio.

You should be able to get 8mA of drive on all bands with the easiest being 160 and hardest being 10. If the drive stops increasing at some point then you might have a soft 5763 driver tube. I found that this tube needs to be in very good condition. Yeah, I know you tried another one...

The lower output on 160 might be due to the RF coupling/D.C. blocking cap in the plate circuit of the 6146s. Stock, there’s a single 500pF cap but it’s become popular to double that up to 1000pF either by adding another in parallel or changing it out.

I don't think a Valiant is ever fully repaired or restored. It just works a while between repairs. They are beautiful transmitters but I'll never have another. Life is too short.

I don't have another in the tube pile, so I'll have to get some. 


That blocking cap is 2000 pF 15 KV on mine.    I will do some more test on 160 tonight.   I can't see how L6C can make much of a difference on that band,  yet the Aux/Fine controls are set differently after installation.   10/1 (now) as opposed to 7/3 before.    We'll see.... 

It’s possible that you’re finals are tuned to a harmonic of the 160M frequency. That band requires most if not all of the available loading capacitance and the AUX control should be on 4 or lower (a bit counterintuitive).

I use AUX 2 and FINE 8 for 1885.

Generally, it’s best to start with these controls on the low side and bring them up to increase loading.

In CW,  there's a huge difference in the coupling between 160 and 80,  yet not so much for AM.   I was using the CW settings as my testing is on CW ( I don't care to zorch a mod tranny if I don't have to......).

By the way,  what does "coupling" stand for (between the Aux and Fine settings on 160M CW ?   I can no find any references to it in the manual.    This might explain the big difference noted above?

"Coupling" and "Loading" both refer to the setting of power transfer to the antenna. For example, minimum amounts would produce a very steep dip in plate current at resonance and low output. Increasing coupling or loading increases the transfer of power to the antenna and the amount of power drawn by the finals. In this transmitter, this is done by reducing the capacitance in the output of the tuning circuit.

The most number of capacitors, greatest capacitance, lowest coupling occur at "0". Lower frequencies require more of this condition than higher frequencies.

"Loading" is referred to in various ways, which can be confusing, which is why the Valiant probably uses "Coupling". There are also historic reasons...

Since CW requires greater power transfer, the settings are higher than they are for AM. The 160/80M CW/AM difference is likely due to the greater amount of capacitance needed for a given change in coupling at lower frequencies.

A frequency counter may show that your 125 Watts is on an 80 meter frequency.

I guess I wasn't clear what I was asking;   Table 4 says "Coupling 1" for 160M,  so it's trying to tell me something.  None of the other bands / mode combinations have this notation.    There's no extra controls around anywhere,  so how do I effect "Coupling 1" on 160M without using the AUX and Fine settings?  I assume they put that notation there for a reason.  Maybe that answer is buried in the manual,  but I haven't found it yet.

Ah gotcha.

It might have been a typo. After typing that whole manual by hand (1957) the person might have said "fooey", or whatever obscenity was in vogue at the time, and left it.

>>I use AUX 2 and FINE 8 for 1885.

I assume that's for AM and not CW?    I'm getting 150W out @450 mA Ip and 8 mA Ig on 1880 CW,  and the inline counter shows 1881.

Tonight I moved the Valiant off the bench and over to the operating position.   I want to use it on the air for a while before deciding what to do next, if anything.

Clark, thanks for all you help!   Stay tuned......

 

Well I've had fun playing with the Valiant on  15, 40 and 80 CW.   

I'm noticing a drift on CW;  not a chirp.   I've done the "Johnson VFO chirp frequency jump" mods as per W8JI.

What I'm hearing is that the TX frequency will change between 70 and 110 Hz over about 2 minutes on CW,  with most of the change occurring closer to the start of a CW transmission.

I'm measuring this with my in-line frequency counter.

I've checked for ground issues etc.   Also, I'm not sure if this is "typical" of a Valiant on CW.

Comments appreciated.

Most likely caused by a rise in temperature within the xmtr due to the average plate currents in the tubes.  Does this drift also occur on AM?

Drift is normal as component temperature stabilizes during the first 30 minutes. Do you get additional drift after this period?  Try setting the Oscillator switch to Zero to see if it drifts with just the low level stages energized.

The drift occurs on AM.   I just tried it on 7290 into a dummy load;  the frequency increased as I keyed the mic (no audio) and stabilized at +100 Hz after 120 seconds.   The Valiant was on for 2.5 hours before this test.

For the CW tests I was in the Novice Rig Roundup as I figured that would be a good place to try CW on a number of bands in a short amount of time,   and the radio was on for hours before I'd start making contacts over several days.

Does the drift occur when the Oscillator switch it turned to "Zero"? This turns on all the low level stages. You might need to put a frequency counter pickup near the driver tank coil.

A number of things can cause that drift.  When you key up you add load to the AC line.  This will cause a slight drop in filament voltage.  This will cause a drift.  The added heat from plate currents will cause a drift.

When those transmitters were made nobody had fancy freq counters to see a 100Hz drift.

In my HB xmtr the VFO is on all the time.  The VFO filaments are run from a separate regulated 6.2VDC supply.

Fred

I ended up putting a multi-loop pickup coil on top of the 5763.

Started out at 7291.34 and ended up at 7291.31/0 after 120 seconds.

Fred, yes I know we didn't have frequency counters back then, but the drift is noticeable on CW.  My 5100B drifts a bit on CW as well but nothing like this, which is why I was curious if this was just what a Valiant did.

I’m still concerned about the modulator resting current.   With R61 at one extreme I can adjust the current to be about 70 volts.    The last time I was on the air I checked and it was more like 75 mA.     

This morning I hauled (ugh) the Valiant back to the bench.

R61 measures 4.9K.    To get 70 mA of resting current, it’s adjusted so that the CT to ground reading is maximum.    With all wires connected to R61 that’s about 4.1K.

I disconnected R15 from R61 and the wire from R61 to T4 CT.

R15 is 15K.   

I measure 522 ohms from the disconnected wire to V17-5 and 467 ohms to V16-5 which goes through T4.

R46 is 101 ohms and R47 is 102 ohms – both OK.

R50 is 960 ohms, which is OK.

I’m not sure what could be the problem as the resistance measurement look fine.  I’m really considering adding a 1K resistor between the R61 CT and T4 CT  (Clark is cringing at that statement….).

Not the optimal solution, but the radio is playing pretty well now......

Well adding in the 1K did nothing.  In fact, adding in any additional resistance didn't do much.   Probably because the current flow through there is so small.

The manual says I should have -46V on pin 5 of the modulator tubes.    I have -49.7 to yield 70 mA modulator resting current.   If I change R61 to get -46V the modulator resting current is 200 mA !

So, maybe change the 15K resistor to a smaller value to put more drop across the pot?   

Leave well enough alone?   The 70 mA is in range... at the top end.....

So there are two issues here: frequency drift and mod bias

Frequency drift: "Started out at 7291.34 and ended up at 7291.31/0 after 120 seconds."  That's with the Oscillator switch turned to "Zero" and the transmitter *not* keyed, right? This is important to confirm.

Mod Bias:  R15?  That's the screen dropping resistor for the RF finals. Did you mean R63? Yes, it should be 15K. Forget the -46 Volts in the manual. The grid current is everything. Check to make sure you have at least -265 Volts at the bias supply (- lead of C93A). I had bad filter caps in a Valiant and that caused low bias voltage.

If all that looks good, gasp, then you could try adding a resistor between R63 and R61 to increase the bias setting range of R61.

Yes.


Yes.


A while back, to address this issue, I removed the voltage dropping resistor from the solid state replacement for the 6BY5G.

This caused the voltage at the - lead of C93A to go from -225V to -240V.    Perhaps decreasing R50 to bring it up to -265V?  I don't see any other way to do that.    The other way would be to leave R50 as is and add a resistor between R63 and R61.

R50 should be left at 1k since that’s part of the “Pi” filter. Either something is loading down the supply or one of the filter caps isn’t right. Is R23 9k? You could also try disconnecting R63 to see if full bias gets you between 50 and 70mA.

I would have also suggested that the shunt wire R59 might be out of spec but the voltage you’re getting from the bias supply is very low, particularly considering today’s high line voltage and the solid state rectifiers.

The next drift test is to compare the bands. 160/80 use a different circuit than 40/20. The two parameters are drift in Hz and % difference between bands.

R23 is 9K and all electrolytics are new.

I should have mentioned that I'm running the Valiant at 115V off of a Variac.

That's ok but I would measure the filament voltage at the 6146 and adjust the variac for 6.3vac regardless of the variac output voltage. Doing this will help get the rectified voltages closer to where they should be.
In my experience....the fil voltages are always low when the line is 115vac.
There is always some loss in the line cord and the wiring in the transmitter.
You probably have already checked the line voltage but again....I would set the 6.3vac fil voltage with the variac.

^^ Agreed. The key is to not have significantly higher voltages than what you find in the manual.

As for the drift, do you have another 6AU6 to try as V1? I've noticed differences in characteristics between brands, and in this critical application, it could solve the issue. Also, have you checked R18 that it is actually 18k? A lower resistance will cause excessive current in V2, the OA2 regulator, which is another tube to check.

Dave and Clark

Thanks for your comments.

I will check what the filament voltage is for 115VAC in.

R18 being 18K?    R18 is supposed to be 20K.   Do you mean R3?    I replaced that with a 10W 20K resistor under the chassis.

Ah! Touche!  Yes, R3...

I always replace R3 with the correct value as well as locating the resistor in the original location.
A contributing factor with the drifting may be that the VFO compartment never comes up to the temperature that the ckt was designed and compensated for.
With R3 outside the compartment....might be too cool in there...so it takes much longer to settle down if ever.
I know that guys install R3 below the chassis but.......
Probably have to test it to see if there is an impact on the drift.
Also as someone else said....replace the 6CL6 and the VR tube inside.
If your VFO box is missing the vertical phenolic bandspread  variable extensions (5) you can get 5/16" plastic caps and snap into the holes to keep the heat in.
I always remove those damn things because the clip thing that fits onto the variable falls off.
To adjust the band spread I remove one cap at a time and stick a RF sniffer in the hole and get the freq right on.
This saves taking that VFO apart to try to get those clips back on....that's a real hassle.

I did some measurements with the line voltage at 115 and 120 VAC.   I used the voltages in Table 1, with the Valiant configured accordingly, as reference.

Executive Summary:   The Valiant likes 120V better than 115 V.

Details:

Notation:    100/80 means 100V HV OFF and 80V HV ON......

Measurements made using a Fluke DVM  (which is probably not 20K ohms/V......)

Bias Reference:  -265V

115V AC:   -226/-202
120V AC:   -251/-228

Modulator Resting Current:

115V AC:  75 mA Min
120V AC:  50 mA Min

Low Voltage Reference: 300 VDC (Measured at J8-4):

115V AC:   348/289      120V AC:   381/320

Power Output:

115V AC:  140W        120V AC:  170W

VFO Heater Voltages:

TB6-2 Reference 5.8 VAC

115V AC:  5.6/5.3  VAC     120V AC:  6.1/5.9 VAC

TB6-5 Reference 5.8 VAC

115V AC:  5.4/5.0  VAC     120V AC:  5.9/5.7 VAC

TB6-6 Reference 340 VDC

115V AC:  348/293        120V AC:  379/324

I will lug the beastie to the operating position and redo my drift tests.......   Standby.....

Stability over 120 seconds was better running the radio on 120 VAC instead of 115 VAC.

40M CW: (330 mA Ip)
+30 Sec:     +80 Hz
+60 Sec:     +90 Hz
+90 Sec:     +90 Hz
+120 Sec:    +80 Hz
Total of 90 Hz over 2 minutes.

40M AM:   (330 mA Ip,  60 mA Im)
+30 Sec:     +20 Hz
+60 Sec:     +30 Hz
+90 Sec:     +40 Hz
+120 Sec:   +40 Hz
Total of 40 Hz over 2 minutes.

I'm going to put the Valiant back in the case and put in on-line.

Great news! Seems like the bias supply issue is resolved as well.

Good news....glad to have helped out a bit with the drifting.