Apache TX-1 HV Short

[oldsalt]

This TX-1 Apache powers up, filaments glow, and no hum nor smoke.  But when switching to the transmit position, the fuse pops and there is an arc flash at what I think is the plate relay  I'm guessing this is a HV short, bad caps, resistors...

Can you help this old appliance op by walking me through the obvious checkpoints.  Have a VOM and lots of ambition to get this rig on Ya AM.

Russ
WA1JFX

[W3FJJ]

First thing is to visually inspect all connections from hi voltage rectifiers, filter caps, bleeder,  lead from power suplply, feed thru insulators, to where the HV is feed to  to modulator, and final.
If you don't see anything that has  arced to ground,  Pull the final/mod tubes and fire up. And see if it still arcs.
If so keep pulling all tubes, even the rectifiers and see if it still blows the fuse..

Let us know what you find..

GL Chuck

This TX-1 Apache powers up, filaments glow, and no hum nor smoke.  But when switching to the transmit position, the fuse pops and there is an arc flash at what I think is the plate relay  I'm guessing this is a HV short, bad caps, resistors...

Can you help this old appliance op by walking me through the obvious checkpoints.  Have a VOM and lots of ambition to get this rig on Ya AM.

Russ
WA1JFX

[WBear2GCR]

I'd work it the other way.

Unplug the rig.
Take your voltmeter and make sure there is NO HV on the plate caps.
Remove ALL the large tubes - that's the mod tubes and the output tubes.
(open cage top for access.)
Identify the HV rectifier, and the LV rectifier.

Replace the fuse.
Remove the rig from cabinet.

Download manual from BAMA (amfone main page upper buttons, click)
Open schematic.

Identify convenient test points for:
1. LV B+ after filter cap
2. LV B+ after rectifier tube, before filter
3. HV B+ after filter cap/pi filter
4. HV B+ after rectifier tube, before filter

A. Fire up AC with NO rectifiers in place - check AC voltages going to the sockets of the rectifier tubes - can be done from the top into the tube sockets.

B. plug in LV rectifier, fire up rig with meter attached to the test point 1.
C. plug in HV rectifier, fire up rig - you may or may not need to key the transmitter to actuate the B+, see schematic, meter per 3.

IF the fuse blows at either point, then you've almost certainly got a shorted filter cap - very common.

With the AC unplugged, check the filter caps with your ohmeter, after checking them with DC VOLTMETER to be sure they are discharged fully. IF they read ZERO or some constant low impedance, they are bad. Replace.

Or remove lead to suspicious cap/filter assembly and test per 2 or 4, as depending on which ur testing.

Most likely a bad filter cap.
Less likely a shorted transformer - like the power iron or mod iron.
On occasion you get a short in a choke - sometimes you can get away with just lifting the choke above ground, if it is a short to the core. BUT YOU MUST LABEL IT CLEARLY AS BEING B+ HOT - and preferably install an insulated shield to prevent inadvertent contact. Best to replace though...

If the power is all good, then suspect a shorted tube.
Plug them in one at a time and see what happens... 

Report back. 

         _-_-WBear2GCR

[WQ9E]

In addition to the excellent advice you have received so far, take a very close look at your 5R4 sockets and the base of the 5R4 tubes themselves.  I have repaired several rigs using 5R4's that had "carbon tracks" on the sockets and tube bases causing fireworks and fuse blowing when the plate supply is activated.  The physical layout of the 5R4 makes it and its socket particularly prone to this given a little bit of dirt and debris getting where it shouldn't.

Rodger WQ9E

[oldsalt]

Much obliged, Chuck, Bear and Rodger.

The rectifier tubes have been replaced with top hats.  Should I unsolder the two red leads on both legs as well as the middle black connection, then measure AC HV?

Removed finals and modulation tubes and still blows fuse.

Russ

[WQ9E]

Hi Russ,

Before you start disconnecting wires, pull the power plug, HV rectifier replacements, temporarily short pin 2 of one of the 5R4 sockets to ground (to make sure the filter caps are discharged) and check resistance to ground from pin 2 of one of the 5R4 sockets (after removing temporary short)-you should measure around 30K.  Have you been trying to start the HV in the CW or AM position?  If in AM and you think the arc is coming from the plate relay then switch the mode to CW and (with power plug pulled) measure resistance from pin 4 of one of the 6CA7 sockets to ground, should be near infinity.  The relay switches the screen voltage to the modulator and a tube problem or a problem with the modulator screen bypass, C-55 would cause this problem. 

Rodger WQ9E

[oldsalt]

Rodger, I removed both legs of the tophats (1N549s x 3 each).  Then with power up, I switched HV on without blowing a fuse.  Voltage check showed over 1KV on each red lead from xfmr (my VOM only goes to 1000 volts).

I checked each leg of tophats and found one leg of 3 has 2 that show no resistance.  The other leg of 3 has equal readings across each tophat (with resistor and cap).

Russ

[WQ9E]

Hi Russ,

OK, I am not quite sure about the exact nature of the replacements in your TX-1.  Does each 5R4 socket have 2 strings each of the 3 1N549 diodes?  When you say no resistance across some of the diodes are they measuring zero ohms or infinity?  In any case, I believe these are likely to be pretty old units and the reliability is far less than the more modern rectifier diodes.  I would seriously consider rebuilding these with modern rectifier diodes such as using 1N5408 units (3 amp, 1KV rating) and available from Mouser at 10 for $1.60.  The 5R4 tube sockets are in parallel so you could make just a single plug in unit using 4 1N5408's in series for each section (a total of 8 ) and leave the second socket empty.  This will give you plenty of current capacity and you shouldn't add any resistors or capacitors in parallel with these rectifier diodes.

Since you are using SS rectifiers you can also disconnect the filament leads from pins 2 and 8 of the 5R4 sockets,  just make sure that the common cathode connection of the two rectifier strings attach to whichever pin (2 or 8 ) that is connected to L-30 (the HV filter choke).  This will keep the high voltage off of the LV transformer 5 volt winding which is a fairly common source of failure in some transmitters (the Halli HT-32/37 series for example).

Although these rectifier replacements may not be the only (or main) problem given their age and the "odd" resistance measurement I would replace them before you go any further.  Of course if you have a pair of 5R4 tubes around you could test it with those.

Rodger

[WQ9E]

Note to self, do not put a closed parentheses immediately after the number 8 or the BB software converts it to a smiley face.  That definitely changes the meaning of a post.

Rodger

[oldsalt]

Hi again, Rodger and AMfone gang.

Diodes that are bad read full scale continuity (0 ohms?).  There are two strings of 3.

You say no resistors and caps are necessary for the 1N5408 diode strings.  Are the new diodes nowadays a self-contained device with built-in resistors and caps?

The 5R4 sockets had been removed previously.  There is a 3-lug strip with a yellow and a gray wire and no other connections; 6 volts across the lugs, so this must be the filament power for pins 2 and 8 you refer to; or, should I look for the secondary winding filament leads.

I don't see the L30 choke.  Looks like it was removed.

Will order diodes and a choke from Mouser and go from there.

In inspecting solder joints and bare wires in proximity to ground, there have been several that were found.  One wire to the mode switch was broken completly off.

Man oh man, this is great to have superlative advice; thanks!

Russ

[W3SLK]

Some other advice from a Apache owner. Try it in the CW position. This eliminates the mod and drive transformers. Also, I would work in the 'Tune' position instead of 'Operate' until you figure it out. Saves wear and tear. Gud luck.

[oldsalt]

Thank you, SLK.  That helps me understand.

Russ

[oldsalt]

What is the rating of L30?  Mouser part #?

Russ

[WQ9E]

Hi Russ,

I imagine L-30 is still there as it is a BIG piece of iron mounted on the chassis top, it is the input choke for the plate supply and there should be a lead going to it from the cathode end of the currently installed rectifier strings.  I don't have the manual with me here at work but it is one of the enclosed large metal units on the chassis top.

If those diodes are reading zero ohms on a low range (RX1) then they are likely shorted, if it it is on a higher range then they may be OK but the one reading different is then open.  In any case, they should be replaced as according to the specs I have they are only rated for 300 mils at 1,200 volts and they are being run right at their current limit in a TX-1 (for comparison the 5R4 tube rectifiers were designed for about 250 mils per section, per tube into a choke input and with 2 tubes in parallel they were rated for 500 mils).  Since they are in a full wave center tap design theoretically you can draw 600 mils continuous from these SS replacement units but that leaves no safety margin for surge and temperature de-rating.

Modern rectifiers from the same batch have very similar characteristics and they should not need any extra components hung on to compensate for differences in switching time or reverse resistance.  Using 4 in series for each string gives you plenty of safety factor for the PIV rating.

If the various electrolytics have not been replaced yet then you might want to add those to your order for Mouser to avoid future failures.  Modern units will be considerably smaller and some of the standard values have changed so for example 47 uf will be the closest to the old 40 uf value.  You can safely go up a bit in microfarad rating to the current standard value; particularly with choke input filters.  You can also safely go higher with voltage ratings.

Rodger

[oldsalt]

Your right, Rodger, replacing paper caps is going to be another task.  How about ceramic caps and resistors?

What do you recommend for replacements for the 125 uf @ 450VDC and 20-20-20 @ 450 VDC filter caps? 

I looked on Mouser to find info on the 20-20-20 can but didn't have any luck.  Apparently, my terminology of the part name and what's in their database didn't match. 
Found the 1N5408 in their catalog with no problem.  Good looking company with lots of inventory; I've got their catalog ordered.

The old diodes must have broken down.  There were 3 readings: 2; 6K; 200K ohms.  I measured +-/_+ across each.

Thank you.

Russ

[oldsalt]

Bear, you had mentioned the following:

Identify convenient test points for:
1. LV B+ after filter cap
2. LV B+ after rectifier tube, before filter
3. HV B+ after filter cap/pi filter
4. HV B+ after rectifier tube, before filter

According to the schematic, the TX-1has a HV xfmr with a secondary of 1600VCT @ 800ma.  There are 3 leads: 2 red, 1 black.  Which is B+, B-, etc.?

Russ

[WQ9E]

Hi Russ,

For the 125 @ 450 volt capacitors I would go with Mouser part# 140-ls121m2w-2240 which are 120 uf @ 450 volt Xicon brand units for $3.14 each.  You will have to be a little creative in mounting these but that should not be a problem, do be careful where the units are in series to keep the capacitor case which floats above ground well insulated from ground (this is capacitor C-1 which is connected directly to the output of the HV choke L-30; the other filter caps (C-2, C-3 have one side connected to ground so that isn't a concern with them).

 I don't know of any new manufacture HV triple capacitor units so you could just use 3 discrete 22 uf @450 volt units to replace the triple.  There should be plenty of room to mount them below chassis and I have used a lot of the Xicon brand units (which are around 60 cents each) with no problems.  I believe there may be some people (Frontier rings a bell) who make "new" multi section units by stuffing discrete caps into old units.  I bought some similar items when I restored my  Collins S line and KWM-2 but for most rigs I generally leave the old unit mounted in place on the chassis for looks and replace the caps below chassis using a terminal strip.

Do be careful replacing the bias filter caps (C-4, C-5) since this is a negative (with respect to ground) output voltage and the positive side of these filter caps must be connected to ground.

Rodger WQ9E

[WBear2GCR]

Bear, you had mentioned the following:

Identify convenient test points for:
1. LV B+ after filter cap
2. LV B+ after rectifier tube, before filter
3. HV B+ after filter cap/pi filter
4. HV B+ after rectifier tube, before filter

According to the schematic, the TX-1has a HV xfmr with a secondary of 1600VCT @ 800ma.  There are 3 leads: 2 red, 1 black.  Which is B+, B-, etc.?

Russ

The B+ is the one going to the input of the filter - a choke, followed by a cap, iirc.

It's a full wave, with a center tap - refer to your ARRL or Bill Orr handbook for the standard wiring configurations for choke input power supplies, and tube rectification for full wave.

You'll see it there!! 

                   _-_-bear

[oldsalt]

Well, the parts are on their way.  Hopefully this will solve the power supply problems.  Here's the list, just in case someone else (newbie) out there wants to tackle a TX-1 project.

1  140-LS121M2W-2240    2 @ 450V 120uF 

2  75-TE1301-E3    1 @ 50V 2.0uF

3  140-XAL50V100-RC  1 @ 50V 100uF 20% 
 
4  140-XAL160V22    2 @ 160V 22uF 20% 
 
5  140-XAL450V22-RC    3 @ 450V 22uF 20% 

6  140-XAL25V10-RC    1 @ 25V 10uF 20% 

7  598-DME6P1-1K-F    2 @ .1UF 630V 10%

8  512-1N5408    8 @ 3.0a Rectifier  (diodes)

Mouser didn't have the .1uf @ 400 volt paper capacitor, so I ordered polyester film caps.  Will those work?

Thanks Rodger and Bear for your recent postings; every bit of info helps.  We've covered a lot of ground in the past 3 days.

I will be careful with HV.  Back in the late 60's, I built a homebrew SSB grounded grid amp using a pair of 813s with a voltage doubler circuit.  Did get zapped a couple of times; I thought I was a goner.  So thanks for the safety reminders throughout our discussions.

I'll report back when the parts get here/installed.  In the meantime, maybe I should check resistor values and study the wiring in relation to the schematic.

Thanks again everyone.

Russ

[WA1GFZ]

Diodes don't wear out it takes a short so find the short before you turn the power on unless you like to replace diodes. If it is a choke change it to the negative lead filter floating transformers can kill you.
Film caps are fine for audio.

[WQ9E]

Hi Russ,

The polyester cap will be fine, I don't think paper caps have been available new for a long time.

I have a feeling that those original rectifiers were probably the problem (given the measurements you made) so hopefully your HV failure will be gone.

I agree with GFZ that floating the choke case above ground is generally a bad idea; if for no other reason you or a future owner will forget about this and touch something that is normally safe to touch.  A potential problem of negative lead filtering was pointed out by Bill Orr in his handbooks in that some transformers do not like having their center tap above ground (lack of insulation) and putting the filter choke in the negative lead will cause this issue.  It probably isn't a problem with most transformers (for example the low voltage units used in receivers often have the CT lifted for standby) but when in doubt I would leave the choke connected as it was originally and replace it if it is defective.

By the way, keep an eye on Mouser quantity discounts when ordering as often times it is worthwhile to order a few extras to hit quantity 10 and receive the discount; particularly for items you will likely need more of in the future.  I remember when I first started collecting and restoring boat anchors I ordered 10 replacement 3 wire power cords and thought I had plenty for a long time; recently I have been going through 30 or more each year.

73, Rodger WQ9E

[N3DRB The Derb]

At the antique radio store I used to work at, we would order 200~300 caps of each value commonly found in old radios at a time. Typical invoice was $1500+ but he got the deep quantity discounts. We went through some caps at that place.

use the old positive leads of the twist-loc caps to solder mount a 90 degree angled term strip of 3 leads, and use the leads to hang your new caps off of. Solder the ring term to the old positive lead of the old cap and disconnect all wires from it. Reconnect to your new terminal leads right above where they used to hook up and reconnect as required, making sure you remember that HV caps in many transmitters of that era are floating above ground,  with the minus side of one cap going to a B- point above ground or chassis
potential. if he can is insulated from ground thats whut ya got. Believe both TX-1 and DX-100 are this way.

[WBear2GCR]

A couple of things...

Modern caps tend to be substantially smaller for the same values than older ones.
You can generally increase any cap that is doing filtering (usually an electrolytic cap) in both voltage rating and value at least 100% and gain benefit and margin.
For small caps, non-polarized type (not an electrolytic) it is a good idea to put in film type caps today. Again, higher voltage rating is often a very good idea.

In an old rig, it's often a good idea to replace all of the filter caps - and all of the signal path caps as well. To keep the freq response the same in the signal path, use the same or very similar value capacitance (ufd), higher voltage rating is ok fine. You can substitute a 0.2 ufd for a 0.22 or even use a 0.3 in 99% of all situations. There are a few circuits where the actual value is critical, but not that many.

It is fine to substitute a film cap for an electrolytic, IF you can find the film cap and it doesn't cost an arm and a leg. For example there may be a 5 ufd. cap used as a bypass on a cathode - a film cap in that slot will actually perform better, and probably last longer (if that matters).

I'll go along with the don't float ur choke - but floating it temporarilly might be a way to make certain that you do have a possible short to ground inside the choke... KEEP UR HANDS OFF FLOATING CHOKES!! 

Btw, those twist-loc caps with the cardboard sleeves on them? Those are the ones that are floating above ground, they have a phenolic mounting base... do not remove the cardboard, and if it is missing, replace it with heat shrink, or tape or something insulated!!

                      _-_-WBear2GCR

PS. one does not have to buy new twistloc caps - you can generally just install regular single caps under the chassis to substitute for the two or three (maybe 4) caps that were internal to the twistloc unit. I've even substituted for just one section that went bad...

[oldsalt]

I found one of the 6CA7 sockets had significant carbon tracks as well as pin 3 (plate to modulation xfmr)) had a cold solder joint.  Pin 4 measures infinity.  Socket on topside of chassis looks healthy; no cracks.

In regards to floating the choke above ground, is there a way to rewire so the case is not hot?  Or should I just wait until I get the rectifier circuit repaired with new parts when those arrive?

Russ

[WA1GFZ]

Just change it to negative lead filtering and the voltage will be only a couple hundred volts of ripple across it and you won't have to float it.