Ampeg Reverberocket 2 GS12-R Power Transformer testing

[guitarblue]

I have been testing the power transformer in a '65 Reverberocket II that once sounded great and then years ago blew a fuse. At some point days or weeks before it blew a fuse it had been turned on without a speaker, but it worked afterward for a while. I put it aside for many years. Now, after it blew another fuse with all the tubes pulled, I finally pulled the chassis, which revealed a fried diode bridge rectifier and a heat-damaged wire or two.

I disconnected the secondaries of the power transformer. All the multimeter continuity and shorts tests between wires and between wires and ground appeared normal to me with what seem to be typical resistance readings. ( including a few ohms between primaries and 25 ohms between B+ secondaries)

Next, leaving all the secondaries disconnected, I first left the old power cord plugged connected and turned the amp on, and it no longer blew a fuse. Unfortunately, I measured zero or less than 1 volt on the secondaries.

I tried taking the switch and fuse out of the circuit -- directly connecting a new, tested 3-prong plug to the primaries, and still I measured zero or in some cases more than zero but less than one volt on all secondary leads compared to ground. The transformer does make a quiet hum.

The black primary lead was attached to the white neutral plug wire when the amp was working, and remained so later when it started blowing fuses. I found this old Rocket 2 Schematic (different amp, same transformer and rectifier, it appears) and if I read the schematic correctly (I am still not very good at that) the black primary is meant to go to the black hot wire from the plug, the reverse of how the two primaries on my amp is wired. Especially because the amp was working before, I am not sure how it could make any difference, or how it could cause the primary to fail to induce voltage on the secondary. I still have not reversed them to see if anything good or bad happens.

Also, I do not understand why a double-pole on-off switch is used, instead of a single-pole switch, and I am not clear what is meant to be going on while it is in the off position with the two black wires on the lower "off" side. (See picture attached) Can anyone explain?

Should I assume the power transformer is dead despite so many normal readings? If so, is this a typical mode of failure?

Do you know what voltage I should have measured on the B+ red secondaries?
Or where to get a decent replacement at a decent price?
It's an Electrical Windings (now TMI) model 550 rated at 110 volts (682-550). Wall current measures 120 volts in my house. I left TMI a message, also Heyboar. Fliptops was out, and I think would have wanted upwards of $150.

I'm pretty new at this - only thing I have restored is a Kalamazoo Model 1. Reading quite a bit. Thanks for any thoughts.

[Stevem]

I did a whole bunch of looking and I can’t find anything that would be a drop in for that.

I would just wait for Fliptpos to get there replacements for a PT-110 back in stock, you might turn up a used on but you would in a short time be right back where you are now.

[R.G.]

I have been testing the power transformer in a '65 Reverberocket II that once sounded great and then years ago blew a fuse. At some point days or weeks before it blew a fuse it had been turned on without a speaker, but it worked afterward for a while. I put it aside for many years. Now, after it blew another fuse with all the tubes pulled, I finally pulled the chassis, which revealed a fried diode bridge rectifier and a heat-damaged wire or two.

That may be the proximate cause - dying rectifiers can kill a power transformer.

I disconnected the secondaries of the power transformer. All the multimeter continuity and shorts tests between wires and between wires and ground appeared normal to me with what seem to be typical resistance readings. ( including a few ohms between primaries and 25 ohms between B+ secondaries)

Next, leaving all the secondaries disconnected, I first left the old power cord plugged connected and turned the amp on, and it no longer blew a fuse. Unfortunately, I measured zero or less than 1 volt on the secondaries.

If you are certain (that is, verifed by measurement) that you put AC mains voltage on the ends of the primary winding and no secondary had appropriate voltages, it's dead.

However, since the PT is the most expensive electronic component in the amp, it is good to be sure.
1. Disconnect ALL PT wires. Use your meter set to read ohms and read the resistance of every wire to every other wire and to the mounting foot. Write this down in a table. It helps if you have labeled the leads before starting so you don't lose track of which is which.
2. Compare the table of what is resistive to what to how the transformer is used in the amp. Each winding - primary, HV secondary, rectifier secondary, heater secondary, etc. - should show low ohms to only its two (or 3, with CT) leads. If any winding shows connection to a different secondary, it's internally shorted, and dead. If any wires are orphans, no other connections, a winding is open and it's dead. If there is exactly ONE wire with connection to the mounting foot or copper belly band, this is an internal screen. Re-check your work and be sure.
3. If the windings all connect end to end and to CT if they're supposed to have one, and do not show shorts to one another, do a pulse inductance test. Get a neon light bulb, like an NE2. hook this across the primary or HV winding. temporarily connect a 6V DC battery across the heater winding and remove it. Just touch it and remove it. If the neon bulb flashes, the transformer is good. If it does not, the transformer is internally shorted and dead. Do this carefully, with all leads open except the ones you're using for the test. Be sure, because replacements are expensive. I can expand on how this works if you are interested.
4. If all that shows good, try what you already did: hook AC mains across the primary and measure winding end to winding end on the various secondaries. Making sure all the transformer leads are NOT connected is important for this. Do not rely on the amp internal connections.

I tried taking the switch and fuse out of the circuit -- directly connecting a new, tested 3-prong plug to the primaries, and still I measured zero or in some cases more than zero but less than one volt on all secondary leads compared to ground. The transformer does make a quiet hum.

That is a bad omen. But do the formal testing above. The ill omen says there is an internal short, but verify that.

The black primary lead was attached to the white neutral plug wire when the amp was working, and remained so later when it started blowing fuses. I found this old Rocket 2 Schematic (different amp, same transformer and rectifier, it appears) and if I read the schematic correctly (I am still not very good at that) the black primary is meant to go to the black hot wire from the plug, the reverse of how the two primaries on my amp is wired. Especially because the amp was working before, I am not sure how it could make any difference, or how it could cause the primary to fail to induce voltage on the secondary. I still have not reversed them to see if anything good or bad happens.

This does not matter. Transformers are not polarized, in this way at least.

Also, I do not understand why a double-pole on-off switch is used, instead of a single-pole switch, and I am not clear what is meant to be going on while it is in the off position with the two black wires on the lower "off" side. (See picture attached) Can anyone explain?

It's a safety thing. Better to break both sides of the incoming AC mains power.

Should I assume the power transformer is dead despite so many normal readings? If so, is this a typical mode of failure.

Do the tests. Do not assume it's dead until you test it.
Notice that you are in extreme electrocution danger while doing these tests, and be very, very careful.

[guitarblue]

Thanks for the advice on testing and heads up on replacement availability.

I notice I previously did not succeed in attaching the schematic of the Rocket 2, which has the same power transformer set-up. I am attaching it now.

Ampeg_gs12r_.pdf

Regarding step 1 you suggest, I took some measurements earlier.
Please tell me the significance of using a transformer foot versus chassis ground. Maybe I should redo the measurements.
I measured small resistances earlier.
Primaries - <4 ohms
Heater filament wires -There was less than one ohm between the heater filament ends, and half that between each and the filament center tap.
High voltage secondaries -- I measured 25 ohms between the red HV leads. There is no HV center tap. I am guessing that is because there is a four-diode bridge rectifier, although I do not know why that eliminates the need.
Because the rectifier is a diode rectifier, there are no rectifier filament wires.
I got infinite resistance between wires belonging to different pairings - concluded no shorts.


I will measure voltages again tomorrow. I am not sure what HV lead to ground is supposed to measure without any load. Maybe 370
or so would be a guess?

Perhaps my meter/leads are acting up or the wire nuts were not doing a good job connecting plug wires connected to primaries . I assume that
100% of leads should be open when measuring voltage - that's what I did earlier. My meter's voltage range up to 500 volts AC. I hope I did not damage it - it still measures mains voltage accurately.

It's late for me and I am scratching my head about this but it sounds like a very good thing to do, so I will try to do it soon.
I guess hardware store still sells those batteries -- my guess is that I will need to order the bulb. Are these the ones?
https://www.amazon.com/Memotronics-Bulb ... 2448&psc=1

"do a pulse inductance test. Get a neon light bulb, like an NE2. hook this across the primary or HV winding. temporarily connect a 6V DC battery across the heater winding and remove it. Just touch it and remove it. If the neon bulb flashes, the transformer is good. If it does not, the transformer is internally shorted and dead. Do this carefully, with all leads open except the ones you're using for the test"

[Stevem]

Face it, that PT is dead so stop wasting your time that you could be putting into the amp elsewhere.

While your at replace all 4 of those diodes with 1N4007 types , if you want less switching noise hash in your audio stream then spring for UF5408 types .
These will cost you about a buck each .
They have thicker leads that need to be dealt with, but they act as a great heat sink so try and keep them as long as possible.
Coiling them up works good.

Whenever I work on a cathode biased Ampeg I always up grade the 10 watt to a 15 watt to get more stability and move that bypass cap as far away from that resistor as I can .

[R.G.]

I notice I previously did not succeed in attaching the schematic of the Rocket 2, which has the same power transformer set-up. I am attaching it now.
Ampeg_gs12r_.pdf
Regarding step 1 you suggest, I took some measurements earlier.
Please tell me the significance of using a transformer foot versus chassis ground. Maybe I should redo the measurements.
I measured small resistances earlier.
Primaries - <4 ohms
Heater filament wires -There was less than one ohm between the heater filament ends, and half that between each and the filament center tap.
High voltage secondaries -- I measured 25 ohms between the red HV leads. There is no HV center tap. I am guessing that is because there is a four-diode bridge rectifier, although I do not know why that eliminates the need.
Because the rectifier is a diode rectifier, there are no rectifier filament wires.
I got infinite resistance between wires belonging to different pairings - concluded no shorts.

OK, I'm collecting my thoughts. I get from this that the primary is not open, the secondaries are not open, and that no windings are shorted to one another.

If you are actually getting AC mains voltage to the primary wires, and no significant high voltage on the high voltage secondaries, you can skip the pulse inductance test.

Given that, I agree with stevem - it's dead. Get a replacement, as much as that hurts financially.

From the schematic, I see that this amp has a variant of the bad old way of wiring the AC mains. It gets correct that the AC mains wire goes first to the fuse, then the power switch. It gets wrong (according to modern practice...) that (1) it's a two-wire line cord, needs to be 3-wire; (2) It has a death cap/switch; (3) it has an accessory power socket. For safety's sake as well as lower hum and noise, you really ought to rewire for a 3-wire line cord and remove the death cap and accessory socket.

[Astronomicum]

Very strange. Not sure what failure would result in the readings being reported. You are measuring for AC on the secondary, correct?

[R.G.]

Very strange. Not sure what failure would result in the readings being reported. You are measuring for AC on the secondary, correct?

An internal soft short would do it.

Inside transformers, the magnetic field forces all turns of wire to have exactly the same voltage across them. A "soft short" is a shorted path with a noticeable resistance in it. This loads the turns involved in the short so much that they have low voltage across them. This prevents the other turns from having any higher voltage. The resistance in the shorting path and wires determine how big the current will be. Soft shorts are different from hard shorts only in how much resistance is in the shorted current path to limit current.

This seems to fit all the reported symptoms.

[guitarblue]

Okay, thanks - a consensus, I'll stop testing the power transformer now, and bank the test-for-internal-shorts info for another time.

Heyboar supplies both Mojotone and Fliptops, who have nothing now.
The (not a tech) handling calls for Mercury Magnetics cound not in the moment identify a replacement for the original 682-550, and said he would get back to me.

Is there a way to spec a power transformer from a schematic if nobody knows the original specs? When asked, I gave the MM guy what was admittedly a guess for open HV value -- 370 volts DC

"It gets correct that the AC mains wire goes first to the fuse, then the power switch. It gets wrong (according to modern practice...) that (1) it's a two-wire line cord, needs to be 3-wire; (2) It has a death cap/switch; (3) it has an accessory power socket. For safety's sake as well as lower hum and noise, you really ought to rewire for a 3-wire line cord and remove the death cap and accessory socket."
3-prong plug added and death cap removed earlier, but I don't know what an accessory socket is to know whether or not it has been altered, but I am concerned.
"...if you want less switching noise hash in your audio stream then spring for UF5408 types."
sounds good - diode leads were very fried, and diodes had melted together!

I have really enjoyed this amp, and a couple of good gig memories from it, but it may no longer be a keeper. I better test the ouput transformer before I I start sinking money into this. I am a newbie and my speed in deciphering schematics and comparing to real amp layout is very slow - kind of a meditative practice for spare minutes.
There is an electronics shop decal inside, I suppose it's the same shop that upped the electrolytic filter capacitor values a bit to 80-50-47-22 versus 70-40-40-20 (seems to be what people do, though now there is a 70-40-40 CE can cap), and in passing I noticed an altered resistor value, perhaps to compensate for filter cap value changes, so there is lots to check if I proceed.

"Whenever I work on a cathode biased Ampeg I always up grade the 10 watt to a 15 watt to get more stability and move that bypass cap as far away from that resistor as I can."
I have more than one Ampeg,and they are too noisy, I would do anything in the power supply section to reduce noise. Sorry to be such a newb, but which component needs to be upped from 10 watts to 15 watts? -- is that a diode spec? The diode spec I remember off the top of my head is 1000 volts and a spec called backward something--or-other that some people upgrade from 1 amp to 3 amps.

[Stevem]

The resistor I am suggesting you up grade is this 140 ohm 10watt.

These do not cause a noise issue unless you have a 120 hz hum sound and your output tubes are red plating .

[R.G.]

Is there a way to spec a power transformer from a schematic if nobody knows the original specs? When asked, I gave the MM guy what was admittedly a guess for open HV value -- 370 volts DC

There is a way. The desired DC voltage listed on the schematic at the first filter cap - 385V. The datasheet says that the 7591 tube produces 32W in Class AB1 with a 400V DC power supply, so this reinforces the 385 value. With solid state amplifiers, the DC voltage is 1.414 times the AC voltage, minus the diode drops and any sag from the transformer internal resistances.

To a first order, the schemo says to get 385Vdc. The AC to produce this is 385 / 1.414 = 272V. Add a little to account for sag (since we don't yet know the specifics of the transformer itself, as we haven't picked one) and 280 to 290Vdc would be reasonable.

The 7591 data sheet says that the max plate current in AB1 push pull is 138ma. The entire rest of the amp lives on the current coming out of the first filter cap, and that is shown on the schematic as being 385-345 = 40v dropped over a 1K dropping resistor, or 40ma. So the transformer has to produce about 180ma total. Roughly.

The high voltage needs to produce 180ma DC; some transformers have their DC out current listed, some only specify RMS current. If you run into an RMS-only specced transformer, you can guesstimate that the RMS current in a full wave bridge from the transformer will be 1.6 times the DC out. So a trannie might be specified as 180 * 1.6 = 288ma rms. I'd go for a bit more than that, say 300ma if rated in RMS current; a minimum of 200ma to 250ma if rated in output DC.

Then you add up the datasheet heater currents for the tubes, and you have a nominal set of ratings to look for.

The physical size will be important. Make good measurements of the size of the original transformer and its available space on the chassis to see if the replacement can fit.

3-prong plug added and death cap removed earlier, but I don't know what an accessory socket is to know whether or not it has been altered, but I am concerned.

It's that two-prong AC voltage socket on the back panel. It's shown on the schematic after the incoming power cord.

I have really enjoyed this amp, and a couple of good gig memories from it, but it may no longer be a keeper. I better test the ouput transformer before I I start sinking money into this.

Odds are the OT is OK. A simple test is to feed 6.3Vac from a heater winding into the speaker output and measure the AC voltage on the primaries. If you get much higher AC voltage on the primaries, the OT is first-order OK. You can leave the amp powered off, pull out the output tubes, and measure from plate pin to plate pin in the sockets while you feed AC into the speaker out. Unplug the speakers for this - don't feed them the same 6Vac.

The diode spec I remember off the top of my head is 1000 volts and a spec called backward something--or-other that some people upgrade from 1 amp to 3 amps.

The diodes they used in the original are listed on the schematic as 800V and 0.8A. Modern diodes are much better. The 1N4007 is a 1000V 1A rated diode and a good replacement. I would use the UF4007, which makes for less AC hum noise, or as noted the UF5408, which are low noise, 1000V, and 3A. It is likely that the old diodes just died and that is what killed the thing in the first place.

[R.G.]

I did some scouting for transformers. I think that the Hammond 270HX could work, although you would have to change the rectifiers. See:
https://www.hammfg.com/files/parts/pdf/ ... 1697661948

The HX has 550Vct under load. If you change the rectifiers from a full wave bridge to two diodes in full wave center tapped configuration, it should give about the right voltage.
It has a 5V winding for a tube rectifier, which you won't need and can tape off the leads and ignore otherwise.
It has 6.3Vct for heaters at 6A. I think this is enough, although I have not added up the filament currents from the tubes in the schematic.
This thing is about US$120 to US$140 depending on how hard you scout deals.

It is possible that a smaller one of Hammond's 270 lineup might work if the physical size is too big. It's a first guess.

Edit: second guess
The Hammond 270FX is a tighter fit, but would probably work.
https://www.hammfg.com/files/parts/pdf/ ... 1697661948

[didit]

Hello --

Perhaps simply order one of these ::
https://www.fliptops.net/catalog/p-1005 ... -reccifier

..

[Stevem]

He would love to, but he found them out of stock.

[jbrew73]

GuitarBlue
I may have the correct transformer out of a Gemini amp. Message me if you’re interested in a used one. I don’t particularly have a need for it and may consider selling it if you don’t find a new one.