Need help with Valiant LV power xfmr

[K1CWS]

Bad filter cap in the bias supply took out the 6BY5 which placed a short across the LV power xfmr. It smoked for a few seconds with an arcing sound till the fuse blew. I have replaced the tube with 1N2007s and new caps. The rig still works and the xfmr only gets pleasantly warm after many hours of operation. But the voltages are not balanced. One side of CT is 8VAC higher than the other. I'm guessing it has a few shorted turns.

So -
 
Question #1: should I be looking for a replacement xfmr?

Question #2 : where can I get a replacement or suggestions for a retrofit.

I know that Harbach Electronics has taken over for Peter Dahl, but is too expensive for me.

I don't care about originality, just want it to work FB.

[W3NP]

Bruce,

The fact that the secondary voltages are not the same across either side of CT does not necessarily indicate a bad xfmr. I would think that if you had shorted turns the xfmr would draw high current and blow the fuse again and at best wouldn't be working fine for hours. The voltage readings may not have been symetrical before your meltdown. I have seldom seen a xfmr that is totally balanced - either resistance readings or voltages.

I would just keep an eye on it for awhile. If I had my Valiant opened up I would check the balance of my LV xfmr for you.

Another reason to have a smaller fuse in the primary of the LV xfmr of these old rigs - something I need to do here.  Smoking and arcing of a xfmr would not make me feel good about it either, but I have had that happen in the past with no further problems experienced.

Good luck with it.

[WQ9E]

Bruce,

Like Dave stated, this is probably not an indication of shorted turns.  The way center tapped secondaries are normally wound you have one side wound close to the core and then the other side is wound on top of the first resulting in more wire used for the same number of turns thus there will be more voltage drop under load in one half of the center tapped secondary.  If you measure the resistance of the transformer windings (without power applied of course) you will measure higher resistance for one half of the secondary.  If you measure both sides of center tap with the load removed then then voltages should be close to equal but under load the higher resistance side will drop further.

The biggest killer of LV transformers in the Valiant is shorted filament leads to the 866A HV rectifiers.  In earlier production runs, Johnson used regular hookup wire which is not sufficient for the HV that does appear on the filament leads to the HV rectifiers.  If you are still using tube type rectifiers make sure that these leads are properly insulated and if you have gone to solid state then you should remove and insulate these leads from the rectifier sockets.

Rodger WQ9E

[Jim, W5JO]

Consider putting a fuse between the rectifier tube and the filter.  Get a terminal strip and put a strand of wire taken from the shield of coax between two terminals.  Put the input on one side and the output to the circuity to the other.  Do this for HF and the LV.  I know you are strapped for space in the Valiant, but it beats having to have a transformer rewound.

[WA1GFZ]

I yanked the bias rectifier tube out of my Viking 2 and installed a fuse socket in the hole so I could put a smaller fuse on the LV transformer. My transformer runs stone cold after putting both 5 volt windings in series with the primary. My line is usually 126 volts AC so it must have been close to saturation because it used to run hot. 6.3VAC Heaters were close to 6.8 volts.  A shorted turn will fry the transformer in short time. I bet Gary the transformer man could rewind it for you if it dies.

[WZ1M]

My two cents worth. Yes, resistance measurments will differ because of the layering. As for the voltage differance, should be pretty colse to same on both sides of center tap. Theres not enough turns in the 1/2 hv secondary to cause a voltage drop of 8 volts in this transformer. Voltages are determined by VPT (volts per turn), not resistance of the winding. I would say, YES, there is a slight problem with the hv winding. Time will tell.
Regards,
Gary

[WA1GFZ]

Gary unloaded voltages should be equal but when you draw current the drop will be greater on the outer winding.

[WZ1M]

But the statement said, 8 volts higher on one side of center tap. That lead me to believe measurments were taken at the ac side. 8 volts differance is not acceptable.
Regards,
Gary

[WA1GFZ]

Maybe it had a shorted turn and it welded itself with a few less turns.
8 volts doesn't sound like a large number of turns. Or as you said someone didn't wind the right number of turns on the transformer. Been a while and don't remember the volts per turn of that size stack. I'm guessing 1/2 volt per turn.

[WZ1M]

For the record. I have wound seven of these transformers in the last year and all final tests, no load are as follows.
HV = 340 volts, each side of center tap.
LV = 225 volts, each side of center tap.
Fil #1 = 6.7 volt
Fil #2 = 5.3 volts
Fil #3 = 2.6 volts
Volts per turn is .474
RESISTANCE MEASURMENTS:
Red to Red = 118 ohms
Blue to Blue = 78 ohms
fil ament = around 1/2 ohm each.
Hope this helps.
Regards,
Gary

[K1CWS]

Gary, I take it that you rewind xfmrs. If so, what would it cost to rewind my xfmr? Would you actually rewind mine or do you have parts to make one up so I could install it faster? My email is K1CWS@ARRL.net

[WA1GFZ]

Gary,
Too bad you don't have the resistance to the center tap to give an idea of voltage drop with current. Based on a guess I think the transformer in question may have been wound at the end of a spool and missing a few turns.

[WZ1M]

Missing turns are not uncommon. I run into this from time to time.. Even Collins 516F-2 transformers with the same part numbers have different turns than a previous one's I have done. Ask me how I know, hehehehehe.
Learned my lesson with those. Put 50 volts ac across the hv center tap and one side of the hv and measure from center tap to the other side of hv. With no load the output voltage should be the same as the applied voltage.
Regards,
Gary

[WQ9E]

Gary,

I am not surprised you find even commercial transformers where the number of turns is not correct from one side to another of center tap.  A funny case (but sad if you were the individual) was a novice friend back in MS who wound a couple of transformers and he carefully measured the amount of wire to do one side of center tap and then used exactly the the same length of wire for the second so he wouldn't have to worry about counting turns.  Unlike commercial transformers he ended up with units that had exactly the same resistance on each side but the unloaded (and of course loaded) voltages were way out of balance.  He was honest enough to confess to his error and learned a valuable lesson in math.

For the original poster, it is likely he has always had this issue with the transformer.  If there was one or more shorted turns in the secondary winding then there should be lots of smoke in a hurry.  If the problem is due to a slightly different number of turns on each side there shouldn't be any significant impact on the transmitter.  Any slight ripple increase should be negligible after filtering.  I wouldn't expect both sides of a vacuum tube rectifier to maintain perfect balance over time (however I do see sellers advertising "matched pairs" of rectifier tubes on ebay 

Rodger WQ9E

[K1CWS]

I have just run a temp ck. After 3 hours of standby operation, the core of the xfmr is 120 deg F. I will take resistance cks after it cools down.

[WA1GFZ]

That sounds about right. My V2 ran about that hot before I added the 5 volt windings in series with the primary.

[K1CWS]

Here are the cold resistance readings: 58.5 - 39.1 - 0 - 40.4 - 61.8

WA1GFZ, you mentioned using 5V winding to buck the LV pri voltage. I thiink that was the fil winding for the 5V4. I guess u changed to SS diodes for the 300 V supply. If so, did u change the filter ckt to choke input? I tried just using SS diodes (1N2007s) and the LV was abt 400V. But if I removed the input cap, it was abt 280V which I think shud B ok.

[WQ9E]

Without the input cap, you probably have considerably higher ripple.  You could offset this somewhat with more capacitance on the output side.  However, I would suggest staying with the standard filter setup and use a series resistor to reduce the voltage back to normal if you are going with a SS LV rectifier.

Have you replaced the 866 HV rectifiers with SS?  The voltage drop across MV rectifiers is very low so you can go SS without being concerned with adding series resistance.  This would leave the 5 volt cathode winding for these rectifiers free which you could then use for dropping the line voltage if needed.

Rodger WQ9E

[WA1GFZ]

I think I did switch over to choke input to reduce the LV supply. I thought it a good idea to take the 5 volt winding off the HV supply. The rig runs cooler with 4 less tubes. The HV is now 760VDC and the finals like that. I mounted an l bracket off the side of the cahssis and added more cap on the HV.

[WZ1M]

Bruce: