SIMPLE TRANSFORMER PROBLEM

[michael_krell]

Ok so here is the thing. i have a tube amp transformer. There is two primary wires two HV wires and two green wires for the heaters. Now, I have 120V on the primary with a ground wire temporarily floating. i have the heater supply floating, and i have the HV wires directly into a bridge rectifier.



NOW i have a DC supply of about 400VDC (without filter caps). I should be able to ground the DC supply to the green wire of the ground plug. Now when i try to do that I get a blown fuse. when i measure the voltage from the ground wire to the DC I get -200 to the DC minus and +200 to the DC positive. I dont understand why there should even be any potential voltage to the ground UNLESS there is something wrong with the transformer.


please help, I know this isnt a tube amp forum, but i dont get answers from other forums and you guys are so great.

[R.G.]

(1) account for all of the wires into/out of your transformer; which hooks to what, by actual ohmmeter readings?
(2)

I have 120V on the primary with a ground wire temporarily floating.

Does that mean there's a green wire coming out of the transformer floating, or green safety ground from the AC line floating?
(3)

NOW i have a DC supply of about 400VDC (without filter caps). I should be able to ground the DC supply to the green wire of the ground plug. Now when i try to do that I get a blown fuse. when i measure the voltage from the ground wire to the DC I get -200 to the DC minus and +200 to the DC positive. I dont understand why there should even be any potential voltage to the ground UNLESS there is something wrong with the transformer.

What you describe makes sense if the HV winding is centertapped with a grounded CT.
(4) are you sure you can do this without getting electrocuted???

[michael_krell]

the primary wires are two black wires that go into the transformer. the green wire i am speeking is from the AC cord. none of the sets of wires are centertapped, but according to the schematic i should be able to do this. This is for an ampeg V4.


ANd yes i can do this without being shocked

[zachary vex]

i'm extremely hesitant to suggest anything that would cause you to continue messing around with a high-voltage transformer that you don't understand, but what RG suggested is very important... disconnect that transformer from everthing in the circuit and measure all the resistances between wires and create a drawing showing the most likely relationship between all of them. if you use a very low-resistance scale, you should be able to figure out where the symmetries exist and guess which wires are center taps to which windings.

like RG suggested in #3, the wire that measures at a center potential between the + and - of the high voltage is probably a center-tap on the HV winding, used typically for a two-rectifier power supply which is not what you are doing (you're using a full-wave bridge which has 4 diodes, not two.) in that case, the HV center tap would be capped off and unused.

but honestly, if you weren't familiar enough with high-voltage transformers to know this before, imagine what else you might not know that could kill you? keep reading about linear power supply design and study schematics until you understand this stuff before you continue working with high voltage supplies!

[zachary vex]

oops, delete!

[michael_krell]

ok seriously, i have a bachelors in electrical engineering, i understand it all. i am just asking a simple question.

[zachary vex]

well, then, it sorta sounds like there's a short from the middle of the high-voltage winding to chassis, doesn't it?

[michael_krell]

i would say so, it sounds like there is. i did check for shorts with an ohmeter, but i didnt find any. and either way i have the case of the transformer isolated from the outside ground, so it wouldnt matter really. so in a strange way i am geting the propper voltages the transformer is supposed to give, but the ground from the AC plug has a hig potential to the DC supply ground.

[puretube]

is it a (cheap/surplus) autotransformer? (prim & sec internally connected)?

[michael_krell]

It is from an Ampeg V4 Head. All of the wires go through a hole in the top of the transformer. i dont know if it necessarilly cheap.

[R.G.]

Well, as I'm so fond of saying, Mother Nature is trying to teach us something. Consider the possibilities:
(1) there really is a high DC potential between AC safety ground and the AC power line; given the average skill of electricians these days, it **could** happen, I guess
(2) there is an internal short of the windings somehow. Could happen. Use your ohmmeter and find out whether the windings are truly and actually isolated from one another by over 1M ohm.
(3) the transformer is slighly fried, and an internal short between windings only becomes apparent when the potential between windings gets high enough; this one is going to be very difficult to find - it's like having a demon inside the transformer shorting something only when you can't look at it.
(4) the transformer designers played clever, putting an internal centertap on the HV winding and connecting it internally to the transformer core, and hence chassis. Works OK in only the intended use. Hmmm... is this thing on a nonconductive surface or bolted to a chassis?
(5) by some oversight, the transformer windings are not actually disconnected from all connections. Possibly a lead wire is shorted to the protective end bells or the copper hum strap inside the bells where you can't see it and only makes contact when you move the leads to hook them up to the rectifiers.

lessee... are there others? Can't think of any.

You are absolutely correct - the transformers they taught you about in class should have isolated windings that will let you derive a supply from them which floats well enough to tie it to AC safety ground. However, there is not enough time in class to teach you about all the perversities in the transformer world.

So the task at hand is to find out exactly what Mother Nature is trying to teach us. All you have is to do this with is an ohmmeter, voltmeter, and your brain; the brain is by far the most important tool, as you know.

I recommend this.
(1) Track down every single lead coming out of that transformer. Make up a drawing, and label ever single one. In this case, the transformer core, end bells and any copper belly band are "leads".
(2) Use your ohmmeter to test continuity in a conbinatorial sense; yep, test every single lead to ever other single lead. It doesn't take all that long if you're organized about it.
(3) Since we don't fully understand what's happening, set things up so that you are not deluged in repairs when you make some mistakes. Either use a 25W 120V light bulb in series with the primary voltage to slow things down a bit and save some money on fuses or conduct your experiments at lower voltage by driving the primary with 12Vac from another transformer and getting results at 1/10 the voltage until you understand better what Mother is saying.
(4) Use plastic wire nuts on all the leads that are not under test at the moment to reduce the possibility of error.

I did not mean to slight your skills.

The profile of the person most likely to ask this kind of question in this forum is a 14 year old who has taken apart an amp or some other piece of equipment to make an unlicensed nuclear accelerator or some such and since he's immortal he doesn't fully understand that grabbing the wrong wire would make him permanently dead. I've been doing this stuff for decades and it still scares me. Maybe more so now that I have less runway in front of me than I did back then.

[michael_krell]

I did say SIMPLE question in the title. hmmm didnt mean to sound like a 14 year old. I unerstand the dangers. i do electrical contracting and have dealt with much more dangerous things, so i know how to work carefully. The transformer is bolted right to the chasis, but the ground is seperated from the chasis all together, as stated before. I am beginning to think that this amp was modified to use a 3 prong cord to reduce noise, but was not completed. Seems to me like the results i am getting are not incorrect, but in fact a grounded cord trying to be used with a transformer without center taps.

If i am wrong inform me please, I was looking at some older schematics and it seemed to me that they were different than newer ones that had a grounded chassis, but they really werent all that different.

I will take the transformer out and take measurements like you asked, ill even take a picture.

[puretube]

It is from an Ampeg V4 Head. All of the wires go through a hole in the top of the transformer. i dont know if it necessarilly cheap.

I thought it was "a" transformer (of unknown heritage), and you were going to use it for a "to be built" ampeg-like-amp...

for such purposes it would be nice to have a good (=isolated) variac...

for the rest I`m with R.G.`s (safety-) instructions - take care!

[michael_krell]

alright here is the picrture of the transformer.



How here is the impedance measurements


BLACK to BLACK:   1.2 Ohms
RED to RED:           9.4 Ohms
GREEN to GREEN:   .6 Ohms

And, all individual wires measure to the case of the transformer are OL, open circuit. so.....

Black to case:     inf
Black2 to caseL   inf
Red to case:       inf
Red2 to case:     inf
Green to case:    inf
Green2 to case:  inf



There are obviously DC resistance measures. I am going to fool around with a 20VAC transformer I have and see what i can do with that.

[puretube]

black to green, green to red, and red to black is infinite, too?

[michael_krell]

oh i totally forgot about that:


Black to:

Red1: 10 ohms
Red2: 1 ohm


I think i see the problem

there should not be any continuity between the wires. Ok so i think thats the problem

just confirm with me so i can order a new transformer. asap

[puretube]

yep, that`s yer prob...
(and that`s why R.G. noted his point 2.)

now you could go and look why prim & sec are "connected" :

maybe it`s only a case of wire-isolation being worn at the rim of the hole
(missing rubber grommet?): screw off the hood (on top in your pic), and look if you can see broken isolation - if so: isolate well and measure again;
if not: windings are probably melted inside due to overload/-heat;
(donate the thing to some transformer-winder-guy, he can repair it...)

[puretube]

BTW: this is exactly a case,
why the yellow/green "earth" (=safety)-wire from a 3-prong mains cable
must be connected directly and firmly to the amp`s (or effect`s) chassis!!!

[michael_krell]

when i take the case of all i see is wires going in and everything is covered by paper which seems to be very hard. i would like to rip it off but i dont think i would find anything usefull.

[puretube]

:cry:  so you have to go the heaviest way: get a new/other one...

was that an amp you played yourself, and then it broke down,

or did you acquire it "as is"?

one thing: don`t let me be your last judge - wait what R.G. has to say, before you take the next step...