[Rob Quay]

Hi Guys,
I’m in the process of assembling an AWA battery eliminator for an AK 30 and have run into an issue. After putting the B voltage section together there’s a short. When I disconnect either transformer secondary (Triad VPS230-110),the short disappears. The PC has been triple checked for incorrect part placement and solder joints checked for neatness. 
After putting together a half dozen AES eliminators I figured this one would be easy. Not so. I’m looking for advice on tracking down the problem. The printed circuit board gets confusing to my 70 year old eyes! Thoughts?
Thanks! - Rob Q.

[MrFixr55]

Hi Rob,

I hear you re the 70 year old eyes. Is the B section full wave w/ center tap, half wave or bridge? Can you post the schematic?


Is this the AWA (Antique Wireless Association) W3NLB-N6GA PCB?

If so, did you install 4 diodes to make a bridge? If so, did you connect 3 wires from the transformer (connect the center tap to B-)? If you did, that is the issue. Depending on the secondary voltage of the transformer and the desired output voltage, either remove the center tap connection or the 2 diodes that are going from the B- / Center tap connection to the 2 legs of the transformer secondary.



Hope this helps.

[Rob Quay]

Hi MrFixr55,
Thanks for your help. There’s no schematic- the best info is online. Google- AWA Universal Battery Eliminator and you’ll find a description of the eliminator, a photo of the PCB and a picture of the completed unit. The transformer for the B voltage is a Triad VPS230-110 configured to use 110v. The center taps aren’t used. 
While troubleshooting I noticed that the supplier sent 1N400H diodes, which I installed. Are they the same as the 1N4007 diodes that are called for? I can’t find a description for 1N400H diodes. Thanks! - Rob

[MrFixr55]

Hi Rob, 


If you bought the board from AWA, they should have provided a schematic or some documentation. You could report this issue to them.

I can' find a 1N400H.  is it possible that it is a 1N4004?  If so, it should be acceptable.  A 1N3001 is not acceptable, as it has a PRV of only 50V.  A 1N4004 has a PRV of 400V so it is acceptable.  I always use 1N4007s for low current supplies.  If in doubt, you can likely get about 20 - 50 of these for about $5.00 via eBay, even from a US based seller.

The transformer secondary connections are interesting.  It seems that the secondary has 2 windings.  It appears that they are put in series and feeding a bridge rectifier.  The connections on the board for this transformer are numbered, likely for the specific transformer called for.  It is important that the correct wires go in the correct holes to get the windings in series and in phase.

With the transformer disconnected and the diodes in place, Check each diode with the DVM on the "---|<--" scale.  There should read about 0.5 (volt drop, this is not ohms) in one direction but infinite in the other direction.  With the meter on the KOhm Scale, check between pins 7 and 12 on the board.  There may be a relatively low resistance, but the resistance should increase, eventually into the 100K to 1 Meg or higher.  If it does not, then there is a short downstream of the diodes. Looking on the underside of the board, 7 and 12 go to the bridge, 8 and 11 connect the 2 secondaries in series.  Any mixup here will either buck the voltage (no voltage but harmless) or short across at least one secondary winding.

Good luck.

[Rob Quay]

I asked the supplier what a 1N400H diode is and they haven’t responded yet. I got 1N4007 diodes today and I think I’ll pull the other ones and replace them one by one,checking for a short as I go. There’s no reading between pins 7 and 12 so there must be a short somewhere. AWA provided an email address for troubleshooting but it no longer exists. Thanks, I’ll keep trying and keep you posted! - Rob

[MrFixr55]

OK, Rob. That is decent news. If you had no reading with the diodes in, then there at least is no dead short.

You mentioned a short at the beginning of the thread. Is that because you blew a fuse, did the transformer get hot or was there just no output?

More later.

[MrFixr55]

OK, Rob, this has become a private war for me. We gotta fix this thing. I think I have some ideas but need more info. Can you answer the following 5 questions?

1. Does the rest of the power supply (A and C voltages) work?
2. Please read off the voltage labels for the taps for the output that are stenciled to the B+ Output edge of
the board. Is the highest B Voltage 180V?
3. What is the value (uF and Working Volts) of C3?
4. Did you get a package insert or instruction sheet for the transformer?
5. Are the wires numbered?

The following is a link to an instruction sheet for this transformer:

https://catalog.triadmagnetics.com/Asset/VPS230-110.pdf




This transformer is essentially an Isolation transformer with 2 primaries and 2 secondaries. The wires are supposed to be numbered. What is supposed to happen here is that Line 1 connects to Line 5, and Line 2 connects to Line 6 on the primary. Line 8 is supposed to connect to line 11 on the secondary. When you connect the primary to 120V, there will then be 240V between 7 and 12. Test this off board. If you get zero volts but the transformer does not get hot, then the transformer is not phased correctly. The transformer will "buck" itself and the net power will be Zero Volts.

If the Highest available B+ voltage is 180V and C3 is a 300 or so working volt cap, then the primaries are supposed to be in parallel and the secondaries are supposed to be in series and must be in phase.

Let me know. Good luck.

Best Regards,

John, MrFixr

[Rob Quay]

Hi John,
1. I haven’t assembled the A & C sections yet. I prefer to make one mistake at a time. 
2. B+ taps are 22/45/67/90/135/180. 
3. C3 is 100 uf, 450 v. 
4. The link you provided for is similar to the sheet packed with the transformer.
5. The taps on the transformer are numbered like the solder points on the board. I just matched them up. 

**My clue that there was a problem showed up with a high reading on the ammeter when I started applying voltage with a variac, followed by a smoking R10 (TWM5J470E). Never a good sign. I pulled the transformer off of the board and will check it out separately. As always- thanks!
- Rob

[MrFixr55]

OK, R10 smoking is a real "smoking gun (pardon the pun!). Sorry, this is lengthy, but I think I know what is wrong here. Spoiler alert. If there is no insulating hardware between Q1 and the heat sink, this is your issue. Read on for the whole story.

The scientific method:

Observation- A working device does not smoke. When smoke is seen, the device stops working.

Conclusion- Devices are filled with smoke. Smoke is the lifeblood of the device. When the smoke vents from the device, the device fails.

OK, we're getting somewhere. I wish that they silkscreened the wire side with the component symbols or silkscreened the component side with the interconnects, but:

Don't bother taking the transformer off. The "divide and conquer" point is R10. There is definitely an issue downwind (on the "load" side) of R10. It is likely that the transformer and rectifier installation is good. I am looking at both sides of the board and can't tell for sure but I think that R10 is after C3. and is between C3 and the Pass Transistor for the 180V Supply and everything that it powers (and it seems to power everything downwind of it).

Try this:

1. Inspect the board for solder bridges, etc.
2. verify that C3 is installed with the correct polarity (reversing polarity on an electrolytic cap causes it to short). The Band with the - signs should align with the white wedge painted on the board.
3. Remove R10, power up and check across C3. If you get about 300V across the cap, and no smoke or heat, then there is a short somewhere downwind of R10. If no voltage at the cap, but 240V between wires 7 and 12 and about 300V between the paired anodes and paired cathodes of the bridge then R10 is between Diode Bridge and the Cap C3. If R10 is after Cap C3, then anything on the regulator circuit of this PS is suspect. If R10 is before C3, then C3 and anything in the regulator circuit is suspect.

4. If the transformer, bridge and cap C3 check out as installed (300VDC across Cap C3, no smoke when R10 is removed) then start checking the rest of it.
5. With R10 still out of the circuit, Ohm between R10 opposite of C3 and the B- terminal. A short may be detected here.

6. Is there an insulator between Q1 and the heatsink, and is there insulating hardware between the screw that holds Q1 to the heatsink? If not, this may be causing a short. Remove the screws that hold the heatsink to the board (These ground the heatsink to the B- plane. Take caution as not the heatsink is now electrically hot. Is the short cleared. Reinstall R10. Does the power supply now work?

Was there any instruction sheet sent with this board? Can you post it or PM it (whatever is a legal way that does not violate Phorum rules). I am not trashing AWA here; they are a great organization. I know that AWA would like to protect the intellectual property behind this power supply, but they really do need to provide more info than they appear to have done.



How much did it cost for all the parts?

This is a rather neat power supply. I have an ARBE III that I bought instead of building a supply due to time constraints. It works great but is a very simple supply.

This one has to be pretty expensive due to the 3 transformers but is a neat concept.

Good luck, keep posting and I will keep trying to help. I do not have a draw program and my way of printing and scanning is broken, but I think that I have a picture in my head of the schematic.

[Rob Quay]

Thanks for the step by step! I’ll follow through with your suggestions.
The heat sink short got my attention right away. To me it seemed counterintuitive to insulate a component from a heat sink. That wasn’t an issue with the AES battery eliminators I built. 
I was going to buy an Arbeii III for use with the AK 30 but was interested in the AWA  battery eliminator. It would have cost less also. Prices have increased since the PCB was introduced. It also took 7 months to fill the parts order. 
Hopefully our back and forth troubleshooting on the Phorum will be beneficial to anyone struggling with the lack of information out there. I’ll keep you updated!

[Rob Quay]

Success! I knew right away that the problem was a short through the heat sink when you mentioned it. After addressing that the B section is working. Another case of knowing just enough to be dangerous! Many thanks. Hopefully the A and C voltage assembly will be uneventful. - Rob 

I will try to keep the smoke in!

[MrFixr55]

Fantastic!! Glad I was finally able to guide you in the right direction. Knowing that R10 was overheating was the key. The good news is that it is unlikely that anything got damaged, especially if you brought the unit up with a dimbulb tester or Variac.

The AES units were eiither a breadboard or the heatsinks were not grounded. In this case, it would be a good idea to install the insulator sand heatsink compound and reinstall, unless you are mounting the PC in an enclosure. That heatsink is at 300VDC above ground potential so one could be bitten real good if they touch it.

[radiorich]

Hello Rob,
good news !

Sioncerely Richard

[Rob Quay]

All of my battery eliminators are in an enclosure, not only for appearance but also the high voltages involved. When the AWA board is completed I’ll be interested in comparing it to the Arbeii III and the AES versions. Do different radios perform better with different battery eliminators? That is the question!
Thanks John for your help and patience. Of course, now there’s the radio restoration that follows. Stay tuned for more questions and updates! - Rob

[MrFixr55]

Hi Rob,

I am glad to help. Some people do crosswords, some play Sudoku. I like to solve real problems. Besides, If I am here, I'm not blowing money I don't have on ePay, buying radios that I have no room for!

This thing may best the ARBEIII. I am happy with mine, but I like the B Supply on this. For one thing, there is a 180V output. You can get 0.7W out of a 71A in the output stage of an A-K 30. Only problem is that you need -40V for C- to do that. The C- section of your supply likely tops out at -22V like the ARBEIII. The ARBEIII may have more guts on the A side. The C side looks about the same, but there is not much current drawn by a C Circuit.

I Have an A-K 30 and a 33 that I have to fix. I have also repaired the AK10, 20, 20C and 35 in my collection. These sets are not hard to figure out. The worst part of a 30 is the same as the worst part of a 35- the crumbling pot metal pulleys and brass bands that couple the 3 separate tuning caps to the single dial. However, you can reach inside and tune it manually. Other issues are the transformers. They tend to open. The P-T156 sold by AES for about $16 apiece. Similar transformers are available from the ARBEIII Guy. These are supposed to be better. The only other 2 things that are a big issue are the grid leak and 2 wirewound resistors between the tuning cap and coil and the grid if the 1st and 2nd RF Amp. These can be bypassed. The purpose of them is to suppress the possibility of the RF amps oscillating. RCA used these in the Radiola 16 and 17 also. The alternative was the Neutrodyne circuit, but they would have to pay royalties to the Hazeltine corporation (right in my hometown, about 2 miles from me when they were still Hazeltine).