Valiant Restoration Issues

[N1BCG]

I've combed through a kit version Valiant that had numerous mods and issues and I've managed to restore it back to its original design. What remains are two pesky quirks that I haven't been able to overcome.

1) Low grid drive on 20M - All the other bands seem to be normal, if not high, compared to the manual's "suggested drive setting", but 20 is the big exception. The 5763 cathode current seems normal when the exciter is tuned, but the drive control needs to be brought up escessively to achieve 8mA and the 5763 cathode current is much too high like that.

Note: I've noticed that the setting of L17 (oscillator tank) plays some role in this although there seem to be several settings that produce greater drive on various bands yet I have not been able to find any mention of properly setting this coil either in the manual or the assembly instructions.

2) Grid current drops - This I've seen on both this Valiant and on a friends'. With the transmitter properly tuned up, switching to GRID shows a steady drop after keying up yet the output remains steady.

After spending MUCH too much time on this, my objectivity is gone and some fresh suggestions are warmly welcomed!

Clark

[WQ9E]

Clark,

The coupling caps from the driver to the final are often responsible for drive dropping, try replacing this parallel pair.  Of course bad final tubes (gas, secondary emission) can also cause this symptom.  The drive will drop with HV on and the tubes producing power but once set during actual transmission it should remain stable.

Recheck adjustment of L17 and L5 (starts on page 9 of the manual). 

[Pete, WA2CWA]

1) Low grid drive on 20M - All the other bands seem to be normal, if not high, compared to the manual's "suggested drive setting", but 20 is the big exception. The 5763 cathode current seems normal when the exciter is tuned, but the drive control needs to be brought up escessively to achieve 8mA and the 5763 cathode current is much too high like that.

Note: I've noticed that the setting of L17 (oscillator tank) plays some role in this although there seem to be several settings that produce greater drive on various bands yet I have not been able to find any mention of properly setting this coil either in the manual or the assembly instructions.


Clark

L17: See page 10 of the Valiant manual, Section 5b, third paragraph for adjustment procedure.

[N1BCG]

I've read through section 5 "Operational Checks" and located the references to adjusting L5 (crystal osc coil) and L16 (11 meter driver tank) but still no mention of L17, which is in the VFO tank circuit, and logically would be mentioned here.

I'm going to try substituting a resistive load in place of the 6146 grid circuit to try to isolate the issue by monitoring the RF voltage developed across the resistor. The parallel 25pF coupling caps will be replaced to keep the DC off the resistor and eliminate one more suspect component.

[Pete, WA2CWA]

I've read through section 5 "Operational Checks" and located the references to adjusting L5 (crystal osc coil) and L16 (11 meter driver tank) but still no mention of L17, which is in the VFO tank circuit, and logically would be mentioned here.

I'm going to try substituting a resistive load in place of the 6146 grid circuit to try to isolate the issue by monitoring the RF voltage developed across the resistor. The parallel 25pF coupling caps will be replaced to keep the DC off the resistor and eliminate one more suspect component.

A lot depends on which Valiant Operating manual you're working with.
1st JPEG: Page 10 is from Valiant manual dated 7/56 - see  Section 5b, 3rd paragraph
2nd JPEG: Page 3 from Valiant Bulletin #1 dated 10/58 - see 3rd line up from bottom

[N1BCG]

Wow, thanks!! I compared my page 10 with yours and the sentence starting with "Adjust L17..." is missing on my copy in section C.5.b. That's quite an omission.

Other notable differences on that page include:

"Plate current should read approximately 80ma (vs 300ma)..."

"Adjust DRIVE control for 4 ma (vs 8 ma)..."

"...tune FINAL dial for minimum current, approximately 90 ma (vs 200 ma)..."

By the dates you mentioned, it sounds like your Valiant is also the earlier (not Valiant II) model, but still. Those are some dramatic differences and we haven't even broached the 52 vs 70 volt RF bias topic!

[AB2EZ]

Reading the recent posts in this thread, there appears (to me) to be some confusion.

A Valiant (either I or II) should, in fact, normally be driven/tuned/loaded so that the three 6146s will draw a total of around 8mA of grid current, and (for AM) around 330mA of total plate current (when the plate tuning is adjusted to dip the plate current).

The lower values of grid current and plate current (at dip), during the adjustments of L17 and L5, apply to the safe operating conditions that Johnson recommended only for the purpose of adjusting L17 and L5.

The manual (Valiant I or II) can be downloaded from the BAMA manual archive.

As for L17 and L5...

As described in the Valiant I manual (also applies to Valiant II) you should adjust L17... with the transmitter's band switch set to 40 meters, and the VFO set to the top of the 40m band... to maximize the grid current (using the drive control to back down the grid current, while you adjust L17, so the the grid current does not exceed 3 or 4 mA). Doing this will maximize the output of the VFO on 40 meters. Note that the 40 meter VFO output is the input to the buffer stage and the multiplier stage... which produce the 20 meter drive to the finals... so adjusting L17 to maximize the grid current on 40m will also maximize the grid drive available on 20m.

As described in the manual, you should adjust L5 with the transmitter's band switch set to 10m, and the VFO set to around 28.5 MHz (after you finish adjusting L17, as above, with the transmitter's band switch set to 40m, and the VFO set to the top of the 40m band)...to maximize the grid current drawn by the final stage on 10m. The purpose of adjusting L5 is to produce a reasonably high load impedance at the output of the buffer stage on 10 meters through 20 meters.

Having adjusted L17 and L5, as above, hopefully you will have enough final stage grid drive on 20m to produce 8ma of grid current in normal operation.

However, since you appear to be having a grid drive problem only on 20 meters (not 15 meters or 10 meters), I would also suspect a bad contact in the 20m position on either the buffer stage band switch wafer or the multiplier stage band switch wafer... or a bad solder joint between the 20m tap wire and the coil of either the buffer stage tank circuit or the multiplier stage tank circuit.

Stu

[Pete, WA2CWA]

By the dates you mentioned, it sounds like your Valiant is also the earlier (not Valiant II) model, but still. Those are some dramatic differences and we haven't even broached the 52 vs 70 volt RF bias topic!

Although I do own a Valiant, I have no clue if it's an earlier or later model. It's been sitting at the end of my workbench for the last 10+ years waiting for me to work on it. It's never been high on my list.

I guess you can call me a collector of vintage manual ephemera.

There is a reason why I posted both JPEG's. The earlier manual does define a L17 adjustment. However, later manuals (like yours) omitted that information. However, in the 1958 Bulletin #1, page 3, we have a clue what to do with L17 if you read the last 5 lines on page 3. They indicated to remove R6 , 470 ohm resistor and C87 .005 cap. R6 and C87 are connected in parallel with each other. One end connects to pin 1 of the 6CL6 and the other end connects to L4, C24, and to the crystal/VFO switch. The bulletin indicates to replace R6/C87 combo with insulated wire. If your Valiant doesn't have the R6/C87 combo, then maybe you should follow the L17 adjustment instructions that are in the bulletin rather then what I showed on page 10 of the early Valiant manual.

[Pete, WA2CWA]

Reading the recent posts in this thread, there appears (to me) to be some confusion.

A Valiant should, in fact, normally be driven/tuned/loaded so that the three 6146s will draw a total of around 8mA of grid current, and around 300mA of total plate current (when the plate tuning is adjusted to dip the plate current).

The lower values of grid current and plate current (at dip) apply to the safe operating conditions that Johnson recommended only for the purpose of adjusting L17 and L5.

The manual can be downloaded from the BAMA manual archive.

Stu

That manual is the first manual they issued for the Valiant. Later Valiant models had component/circuit changes and some procedural changes in the initial setup/adjustments.

[N1BCG]

This is so frustrating.

I put an RF diode on the grid of the 5763 driver/multiplier and the voltage was amazingly consistent on 40, 20, and 15 yet taking a sample from the plate circuit showed significantly less voltage on 20. I also disconnected the parallel coupling caps to the finals to be sure that wasn't presenting a lower impedance on 20.

And just to add another bucket of confusion into the mix, I noticed that a friend's Valiant, which was factory wired, didn't even have a wire on the 20M tap on the buffer tank coil...the red one...it's just not there and there's no sign of solder having ever touched the small loop that makes the tap. He gets plenty of drive on 20M.

[AB2EZ]

If there is a strong enough signal on the grid of the 5763,  then the weak 14.xxx MHz signal at plate of the 5763 strongly suggests that there is a problem with the 5763 output tuned circuit on 20 meters.

1. Verify that the band switch wiper on the wafer that controls the 5763 output tuned circuit is making a good contact (in the 20m position) to the terminal to which the 20m 5763 output coil tap wire is connected. Make sure that the 20m coil tap wire is making a good, low resistance contact to the coil. (The fact that the solder joint looks okay is not sufficient to ensure a low resistance connection).  You can do this by temporarily disconnecting the 20m coil tap wire at the end where it connects to the band switch wafer (or cutting it in the middle).

2. Verify that the taps on the 5763 output tuned circuit coil are connected to the correct solder terminals of the correct wafer of the band switch.

3. Verify that there is no solder bridge (or other extraneous connection/conducting path) between the 20m solder terminal on the band switch and adjacent terminals on the band switch.

4. Verify that the 20m coil tap wire isn't touching any of the other coil tap wires.

Stu

 

[N1BCG]

Checked all of those in addition to the cap to ground on the "low" side of the tank as well as the feedback cap from plate to grid.

[AB2EZ]

Make sure that inductor L6C is installed properly on band switch wafer 3A, and check that there is no short circuit (solder bridge or wire bridge) across the wafer solder terminals where L6C is connected.

Stu

[AB2EZ]

Important addendum

Coil L6c is wired between the top of L6a and the bottom of L6b. Also terminals 4 and 5 of switch 3A are wired to their respective ends of L6c.

If the wires from the top of L6a and the bottom of L6b are reversed (wired to the wrong terminals on switch 3A)... then you will have too little inductance on 20 meters (L6c will be bypassed on 20m)... and 20m will be the only band affected by this error

Stu

[N1BCG]

Now the questions are really piling up... I looked carefully at my friend's factory wired Valiant to determine why the red wire from the driver tank wasn't ever used. As it turns out, his has the "L6C" (12T 3/8" dia) you mentioned added to the bandswitch yet neither his nor my schematics show that. Even the assembly instructions show just L6A and L6B.

What this is starting to sound like is that my rig and documentation precede a design change that Johnson made at some point. I'm going to fabricate an L6C and give it a try. Perhaps this will resolve the grid drive issues by changing the L/C ratio. Sure hope so.

[AB2EZ]

Okay

It appears that the Johnson engineers decided that they needed a higher Q on 20m.

Be careful to add in the additional coil by exactly following the wiring approach in the BAMA schematic.

Note also that if L is too low on 20m, then C has to be higher... and if there isn't enough C available in the driver/multiplier tuning capacitor, then you would not be able to resonate the tuned circuit at 14.xxx

Stu

[N1BCG]

This is the older Valiant (not Valiant II) and it seems to match the schematic I've found perfectly, yet there is this low grid drive issue on 20m. Compared to my friend's similar rig, except for being factory wired, the inductance used for 20 and 15 are derived very differently.

On mine, all the taps on L6A (mounted behind the oscillator assembly) are used with a red wire for the 20m tap. The only other inductor is L6B with the center tap. This matches the schematics in both our manuals, however...

On his, the L6A tap that would have the red wire has no connection (and never did), and the bandswitch is wired so that the 20m setting utilizes the extra L6A inductance that would have been used for 15m. The addition of L6C (which I can't find any documentation for yet is present in my friend's factory Valiant) is combined with L6B for 15m operation.

I'm hopeful that the addition of an "L6C" and the rewiring of the bandswitch will yield greater drive on 20m through the reduction of driver tuning capacitance.

[AB2EZ]

Don't forget to check the adjustment of L5.

If the transmitter's band switch is in the 20m position, but if L5 is not adjusted properly (too little inductance) then the strong RF signal on the grid of the 5763 might be dominated by a component at the 3rd harmonic of the VFO (21.xxx MHz), and may have only a tiny component at 14.xxx MHz, and, also, only a tiny component at 7.xxx MHz. In that case, the 5763 won't be able to produce the desired 14.xxx output signal.

Unless you use a scope, with a sufficiently high impedance probe to look at the waveform of the signal on the grid of the 5763, you can't assume that measuring a large peak value of the RF component (with a fast RF diode detector) is sufficient to conclude that the input to the 5763 is okay on 20m.

Stu
 

This is so frustrating.

I put an RF diode on the grid of the 5763 driver/multiplier and the voltage was amazingly consistent on 40, 20, and 15 yet taking a sample from the plate circuit showed significantly less voltage on 20. I also disconnected the parallel coupling caps to the finals to be sure that wasn't presenting a lower impedance on 20.

And just to add another bucket of confusion into the mix, I noticed that a friend's Valiant, which was factory wired, didn't even have a wire on the 20M tap on the buffer tank coil...the red one...it's just not there and there's no sign of solder having ever touched the small loop that makes the tap. He gets plenty of drive on 20M.

[w1vtp]

 Regarding Stu's comment: You really need a scope at this point.  You will be looking for ~ 72 nS sinewave for a 14 MHz signal at the test point of interest.  Obviously, a 10:1 test probe should be used, AC coupled,, hi Z


Al

[N1BCG]

You both make excellent points about the scope. A Tektronix 2445 (150 MHz) was used the week before last and all it proved was that the signal was what it should be on 20m but at a greatly reduced level.

Last week was spent testing resistors and capacitors to insure that they were within tolerance and that the kit version Valiant was properly assembled according to an assembly manual and properly adjusted according to a user manual.

The driver tank circuit was checked against Valiant schematics from that rig *plus* two others and what was wired by the original builder was absolutely correct.

And that's why it wasn't working...

A friend's factory-built Valiant had what Stu referred to as "L6C" which was used in place of one of the L6A windings but was NOT in my rig nor mentioned in any documentation. Having exhausted all ideas and noting how well the factory Valiant worked, I wound 12 turns on a 3/8" dia and rewired the band switch with my own L6C and now the drive on 20 meters is more than sufficient.

It certainly would be nice to understand what's going on here.

[AB2EZ]

It's nice to see that you finally have your Valiant working properly on 20m.

When the 5763 output tuned circuit is adjusted for resonance, how is the tuning capacitor positioned... between fully unmeshed and fully meshed?

Did you, by any chance, notice where it was set, for maximum drive on 20m, before you rewired the band switch and added L6c?

You might not be able the judge the setting from the front panel knob position, unless the alignment of the knob and the capacitor's rotational position is known to be accurate.

I'm wondering if there was too much inductance or too little inductance to achieve resonance of the 5763 output tuned circuit on 20m (taking into account the input capacitance of the 3 x 6146 output stage, stray capacitance, etc).

Too much inductance, before the band switch wiring change and the addition of L6c into the overall circuit (even though it may seem, at first glance, like inductance is being added) is a possible source of the problem. I.e. not being able to make the total tuning capacitance, including the input capacitance to the 3 x 6146 output stage and the stray capacitance, small enough to achieve resonance.

Stu

[Pete, WA2CWA]

Where did the information for adding this "L6C" come from? My three different Valiant manuals doesn't have it, Clark's manual doesn't have it, Valiant manual addendum's/bulletins (at least the ones I have) don't have it either. I looked at my Valiant on the bench and it doesn't have it, and when it was running, it ran fine on 20 meters.

[AB2EZ]

Pete

See the BAMA Valiant II manual.

The Valiant II is essentially a Valiant with a cosmetically updated enclosure.

I assume the the Johnson product manager used the introduction of the Valiant II as an opportunity to fix a few problems with the Valiant.

The addition of L6c is almost certainly a "bandaid". There was something problematic about using a small portion of L6a and all of L6b as the 20m driver tank coil. Maybe the problem was dependent on some details of the placement of L6a and L6b. Maybe the problem had something to do with the placement of the wires running between these coils and the band switch. Note that L6a is adjacent to tuning capacitor C7, and, on 20m, the small, un-bypassed section of L6a (in the original design) would have a large ratio of diameter to length. This may have produced problematic interactions between the magnetic field of L6a, ground, and C7.

Whatever the problem is/was, Johnson solved it by adding a "patch" in the form of L6c.

Stu

[ka4koe]

I've got a weird one.....on 15m it takes the grid current a few seconds to go up to 8 mA. Thermal related??? Perhaps bad cap or weak tube. Not a biggie....just something else to fix when I feel like removing the screwless cabinet from the Beast.

[Pete, WA2CWA]

Pete

See the BAMA Valiant II manual.

The Valiant II is essentially a Valiant with a cosmetically updated enclosure.

I assume the the Johnson product manager used the introduction of the Valiant II as an opportunity to fix a few problems with the Valiant.

The addition of L6c is almost certainly a "bandaid". There was something problematic about using a small portion of L6a and all of L6b as the 20m driver tank coil. Maybe the problem was dependent on some details of the placement of L6a and L6b. Maybe the problem had something to do with the placement of the wires running between these coils and the band switch. Note that L6a is adjacent to tuning capacitor C7, and, on 20m, the small, un-bypassed section of L6a (in the original design) would have a large ratio of diameter to length. This may have produced problematic interactions between the magnetic field of L6a, ground, and C7.

Whatever the problem is/was, Johnson solved it by adding a "patch" in the form of L6c.

Stu

I looked at the Valiant II manual on BAMA. The schematic was revised probably to add this coil in 4/63. Most of the parts list is also dated 4/63 but this coil L6C is not listed. My Valiant II manual dated 11/62 doesn't show this L6C coil. Also, going back to the initial L17 adjustment, neither the BAMA Valiant II manual nor my Valiant II manuals document any L17 adjustment. In looking at my Valiant on the bench, the slug in L17 is turned all the way out which seems to agree with my second attachment that I posted earlier.

What I find strange is that from 1956 to 1961 they built 5977 Valiant (I's). They also issued several addendum/bulletins against the Valiant during that time and revised the schematic at least twice and never addressed this addition of a L6C coil. It wasn't until they were into production of the Valiant II, early 1962 with a build of 634 transmitters, that in early 1963, they added this L6C coil. I suspect there was something in the redesign of the Valiant II that initiated this L6C coil. If this was a major problem with Valiant I's running on 20 meters, we would have heard about it long ago and the Johnson engineers would have addressed it much earlier. However, since Clark added a homebrew L6C to his Valiant and it solved his problem, that really all that counts.