My new donar amps here, how to test an OT

[Bob-I]

The seller represented this exactly right, low volume distorted sound. He believed it was an OT but without replacing it, how can I test it? I checked all voltages, I think next I'll inject a sine wave and connect up the scope.

Any ideas?

[skyboltone]

The seller represented this exactly right, low volume distorted sound. He believed it was an OT but without replacing it, how can I test it? I checked all voltages, I think next I'll inject a sine wave and connect up the scope.

Any ideas?

I prefer a megohm meter (megger). Check each winding to ground (case) and primary to secondary at a minimum of 1000VDC, 2500VDC better yet. Look for probably a minimum of 50 megs resistance. Probably more. A new transformer will show greater than 999 megs resistance. An inservice unit would be usable at a lower test. We look for over 10-20 megs minimum on large in wet service motors. This is a somewhat distructive test so don't do it too often.

Get one off EBAY

[Bob-I]

The seller represented this exactly right, low volume distorted sound. He believed it was an OT but without replacing it, how can I test it? I checked all voltages, I think next I'll inject a sine wave and connect up the scope.

Any ideas?

I prefer a megohm meter (megger). Check each winding to ground (case) and primary to secondary at a minimum of 1000VDC, 2500VDC better yet. Look for probably a minimum of 50 megs resistance. Probably more. A new transformer will show greater than 999 megs resistance. An inservice unit would be usable at a lower test. We look for over 10-20 megs minimum on large in wet service motors. This is a somewhat distructive test so don't do it too often.

Get one off EBAY

Interesting, I've learned something.... only these meters cost more than an xformer.

How's this for a plan. Inject a sin wave, check with the scope that 1/2 sine wave is entering the output tubes and that a distorted sine wave is exiting. If I either see a distorted sine wave at the PI output (op amp pi on this amp) or if the wave matches going in and out, then the problem is elsewhere.

Whatcha think?

[otterpop]

If you'll ask your local electrician nicely, he may test it for you. Some large tool rental places may also rent Meggers.

[Icetech]

hook the OT upto a known good amp for a few minutes and see how it works?

[skyboltone]

I'm with you here. But you say it's got low volume right? So you got to take amplitude into account. That's a little tougher.

How about this. Disconnect both primary and secondary windings and using your ohm meter, check for things that make sense. Like on the primary side you have x amount of ohms from plate lead to plate lead and x/2 plate lead center tap. Like on the secondary side you have x ohms from the grounded lead to the 4 ohm tap, maybe around twice that to the 8 ohm tap, (or is it half?) anyway it's a factor of two. Then check both windings for shorts to the case. Your voltohm meter should not be able to detect any ground on the primary or secondary leads. Finally there should be no continudity between primary and secondary windings whatsoever. If there are resistances that just don't fit with what you expect then expect it's a bad OT. Also, pull the power tubes and lay some big DC+ on the primary windings with the case and secondary windings isolated from the chassis. Check for leakage voltage on the case (Red probe on case, black probe on chassis.) I'm not real sure how much this will prove. Say there is DC voltage on the case, put a 200K resistor between the chassis (B-) and the case. This is a bleeder. Then measure for DC on the case if it's gone it would indicate a high impedance leak. Not good certainly but maybe not the source of the problem. If the 200K resistor smokes and stuff, then I expect it's more serious. Just kidding. If there is still DC on the case the OT is toast.

Back to the scope. I'd start by yanking the power tubes. Put the sine wave in there then follow the AC signal from the pre tubes all the way through. You should see the gain at each stage. Right on the plate pin. With everything nickled we shouldn't see much distortion I'm thinking. Some, not a whole bunch. If you inject a 300mv sine wave maybe you got 5 volts at the plate of V1? By the time it gets to the PI we should be seeing maybe 15-20 volts nickled?? Is the PI a unity gain stage in these amps? Don't forget, you guys have way more amplifier experience than me, I'm just thinking out loud through the process. If the PI is a unity gain stage we should see the 15-20 volt half wave coming out too right? Put the power tubes back in. Then if we got good series grid resistors and the screen parasitic suppressor resistors aren't smoked then we should see a really big AC signal on top of the plate supply voltage unless the tubes are bad or the OT is keeping the tubes from working. If the OT is bad I'm thinking the power tubes gotta be suffering too. Are the plates glowing? As you follow the signal through you won't have to change the frequency on the scope just the voltage. Use a good quality scope probe too. For this stuff a scope is for sure a great tool but so too is a VTVM. Remember you are looking at the AC voltage here not the DC, but it's a good bet before you start to check the DC voltages at each stage at the plates right? All that will take maybe 20 minutes. If everything looks good in the amp and you've got low amplitude coming out I'm betting the OT is no good. The OT is presenting a load the tubes can't handle. You can poke around the power tubes with the scope too to see if you've got some parasitic way up there. Just keep turning up through the time base and see if something shows up as you go. Depending on your scope speed a parasitic may be out of range, though. If you've got a little neon bulb, tape it to a chopstick and wave it around close to the power tubes and sockets. If it glows you've got a parasite.

Have fun. I hope this one turns out like yer pals bass amp.
Dan

[heisthl]

Be sure and use your nose - OTs have a distinct smell when abuse has taken place. Most times you can find a bad one by just using an ohm meter and as suggested above, the plate windings should match to the center tap, the secondary should be twice the resistance on the 8 ohm tap as opposed to the 4 ohm and nothing should measure to the case. If you find one of the primaries is open to the center tap and you're going on the cheap, a lot of times you can repair the transformer by opening it up and peeling back the outer paper - the break is usually near where the leadwire attaches to the windings. A little "micro surgery" and you might revive the dead. Sometimes even if you can make the repair there are enough shorted turns inside it sounds crappy anyway.

[Zippy]

I prefer a megohm meter (megger). Check each winding to ground (case) and primary to secondary at a minimum of 1000VDC, 2500VDC better yet. Look for probably a minimum of 50 megs resistance... This is a somewhat distructive test so don't do it too often.

Insulation checks with a megger need not be "somewhat destructive" if you don't exceed the breakdown voltage of the insulation. 2500VDC seems excessive to me. We used to use 500VDC for all the 450VAC motors in the good ol' US Navy - part of the regular preventative maintenance cycle. If the insulation is marginal - yet still good - going overvoltage can kill it by arcing through the varnish.

[Bob-I]

As soon as I saw "somewhat destructive" I thought... nope... not this bozo. I'd kill it for sure.

So I went to plan B. 1st... smell. yuck, the whole amp smells but not of burnt xformers, next, substitite. I connected up a Hammond xformer I have here, no change in sound, still weak and distorted.

Conclusion? somptin's fried but it's not the OT. On to Dumble planning.

Thx everyone, great discussion.

[skyboltone]

I prefer a megohm meter (megger). Check each winding to ground (case) and primary to secondary at a minimum of 1000VDC, 2500VDC better yet. Look for probably a minimum of 50 megs resistance... This is a somewhat distructive test so don't do it too often.

Insulation checks with a megger need not be "somewhat destructive" if you don't exceed the breakdown voltage of the insulation. 2500VDC seems excessive to me. We used to use 500VDC for all the 450VAC motors in the good ol' US Navy - part of the regular preventative maintenance cycle. If the insulation is marginal - yet still good - going overvoltage can kill it by arcing through the varnish.

Hi Zippy, 1000 volts may well be enough for an OT. I dunno. The AC signal is on top of the DC voltage which for most of these tubes is just under 500VDC. The test voltage used by most manufacturers is 2500VAC. This as you know puts a 4325 +/- peak voltage on the insulation. I don't think 2500DC is way out of line but it would be the appropriate megger check on a PT.

YMMV
Dan

[Zippy]

Thanks for the update and insight. Much appreciated.

George