[Skip]

I am working on my 90 with PP 45's.  All seems good but it looks like the 2.5 filament section for the 24 and 27 tubes is shorting out (section A).  My question is, can I just get a filament transformer rated at 2.5CT 10amps and wire it in as a replacement, or do I need to try and find a complete transformer.  The Philco part# is 4938 and I do not think it was used in any other models. 

Transformer I am looking at http://www.mouser.com/ProductDetail/Hamm...2EniHK8%3d

Schematic http://philcoradio.com/tech/images/90a.jpg

Thanks

[Mondial]

If the filament winding has a short, you must replace the power transformer. Even if you disconnect the winding from the tube filaments, it will draw excessive current and cause the transformer to overheat.

Are you sure the transformer filament winding itself is shorted? Have you disconnected it completely and checked the open circuit voltage? It is unusual for the filament winding itself to short. More likely the winding leads where they exit the transformer casing are shorting or something external in the wiring.

[Skip]

Well here is what I have tested, maybe I am looking at this wrong.  With the 80 and 45 tubes out, the voltage on the filaments is 1.4 and the amps are around 8.  With all the tubes out the voltage is 2.6 and the amps are 0.  I checked each tube socket to ground and all along the filament wires for a short.  The ohms on the filament section is .1.  It does take about and hour for the the transformer to get hot to the touch.  

I have not pulled the transformer and opened it up, but I will.

[morzh]

Skip

How is the general power consumption and the non-fil voltages?
If you have a turn short, especially in fil section, you will see other voltages degradation plus large current draw on 110V side. And the 2.5V being 2.6V with no load is a good sign.
If not, look for problem elsewhere.
Your tubes might be shorted.

[Skip]

Well I am stumped.  I have removed the transformer completely.  I took the shields off, checked all the connections, one of the filament leads looked a little loose, so I resoldered it.  I put the transformer back together, checked all the ohms and unloaded voltages. I then took two 27 tubes, and four 24 tubes and wired the filaments up per the schematic with jumpers on my bench.  I wired them to the transformer and powered it up.  The filament voltage unloaded is 2.7v and with the 24&27 tubes hooked up ONLY, the voltage drops to 1.5v.  The current draw is 8amps.   Below is the voltages on the leads with and without the filaments of the 24 and 27 tubes hooked up.  

With 24&27 tubes hooked up only
Pri: 115vac.    .3 amp draw
Sec HV: 732vac
45 tube filament: 2.7vac
80 tube filament: 5.4vac
24&27 tube filament: 1.4vac

Without 24&27 tubes hooked up
Pri: 115vac   4.0ohms

Sec HV: 737vac.  237ohms
45 tube filament: 2.7vac.   .2 ohms
80 tube filament: 5.5vac.   .4 ohms
24&27 tube filament: 2.7vac.   .2ohms

[Radioroslyn]

I had a bad heater winding in my HRO power supply. A few things I noticed. Transformer got hot, after it was running for 15 mins you couldn't touch it. Smelled like it was burning. Heater voltage would vary from 3v up to the normal 6.3v. Had some 60cy hum that could be quenched by connecting an earth ground but there was a fair amount of sparking when connecting the ground wire.
If you do find that it's bad it may be cheaper to have a custom one made or rewinding your old one as Hammond are costly. Unfortunately with a set that has 2.5v tubes it really limits the possibilities for replacement. A transformer out of a 70 has the voltage but I don't think it will make enough current.

Terry

[morzh]

Skip

Measure the filament ohms in all tubes, and also keep taking them out one at a time, see if the draw drops from 8A down. Not sure but might be a bad tube.

[Skip]

Well I have checked all the ohms in the tubes, they are around .1 to .2 when cold.  As I took a tube out the ohms slowly dropped by roughly 1.8 to 2.0 ohms per tube.  It is what I would expect since the filaments of the 24 and 27 tubes draw about 1.7 amps each.  Voltage also slowly went up as the tubes were pulled out of circuit one by one.  I have a few dozen 27 and 24 tubes and have tried roughly a dozen or so.   All with the same results.  When I had the transformer in after recap, the radio played for couple of weeks on my test bench for a few hours.  Then one day while listening to it it just faded away, then after a few mins would fade back in.  I kept watching it and noticed the filaments would dim as it faded away.  That is what made me look at the filament circuit and then the windings.  Also I noticed the transformer after an hour or so would get hot.  But so far while just testing the transformer out of the radio, it has not gotten warm.  

Could it be the filament windings are ever so slowly failing, but I cannot detect a short yet.  It just cannot hold the volts as the load goes up.  

[morzh]

I dont get it. When you say it draws 8A, it is on the filaments side right? You measured current in series with filament winding? In order to draw that much you cannot have shorted turn. Shorted would 1) make the xfmr draw power on its own with no load, 2) would bring all voltages down, including its own winding so your tubes would draw very little if anything. Not the whooping 8A.
Shorted turn acts like a shunt, drawing current fir itself and starving the load, not adding to it.

See if your tubes have phil short to other electrodes.... I dont know but something doesn't jive here.

[gregb]

The transformer could have a winding or two shorted internally because with the six tubes he has connected he should be seeing 10.5 amps at the rated 2.5 VAC and he is only seeing 8 amps. Seems like an internal fault in the transformer to me.

Gregb

[morzh]

Could be. Yes it shoud be 10.5A.
The other voltages though don't seem to be affected. Strange.
Cold filaments will load it but then they should warm up and the load should be OK.
Well, could be the tranny....before tossing it I would exhaust the other possibilities.

[Steve Davis]

The voltage dropping as you increase the load, sounds more like a high resistance rather than a short. I would try re-soldering the leads where they connect to the windings.

Steve

[morzh]

Yes, sounds plausible.
He said though he resoldered something in the xfmr.

The leads from the very first tube where the xfmr connects, that could be bad spot there.

I am still not convinced the xfmr is bad.

[gregb]

Could be a bad solder joint near the beginning of the filament string as well. I might look at the first connection the transformer leads make in the chassis.

Gregb

[Mondial]

Yes, it seems more like a bad connection somewhere rather than a shorted winding. When you are drawing 10 amps of filament current, it takes only a tenth of an ohm of resistance to drop the voltage to 1.5VAC.

If the no load voltage remains at 2.6VAC and is constant then the winding is not shorted. If it drops with load then there is some series resistance somewhere in the connections causing the voltage drop. 

If the winding is not shorted but had an internal intermittent connection, then you possibly could use an external filament transformer to feed the filaments instead. That would be assuming that all the other transformer voltages were correct and stable.