Help with 1965 Vox solid state amp, please!

[AmpRX]

I've been a tube amp guy since I bought my first amp in 1965 (Ampeg Jet). I recently acquired a VOX AC30L (Lightweight). It's rare with three 10" Elacs. The amp is based on the VOX T60 bass amp with an additional Vibrato channel which is housed in a seperate chassis. The amp has low power with only 1.4 volts at the speaker with four or eight ohms. If it was a tube amp it would be no problem to figure out but I have zero ss experience. Normally I would check voltages and then trace the signal from the input to the output. The voltages look fine. I replaced the power amp transistors. Is tracing the signal the same on a ss amp? Input signal to a transistor a certain strength and higher at the output? What signal strenght should the signal be at the input to the power amp transistors? (on a tube amp around 25VAC). I get different readings at the Collectors of the power amp transistors. Is that normal? Can you help? Where can I go to learn more???

You can see the T60 schematic here:

Vox Amp Heaven Schematics - Berkley Cambridge Pacemaker Foundation Defiant Conqueror Escort Super Midas Busker Vigilante

Thanks!

[Tom Phillips]

.. Where can I go to learn more?

One place would be R.G Keen's site at New Page 1 Then search on his site for articles related to "Vox", "Thomas Organ" etc.

[AmpRX]

Thanks. This amp was UK made by JMI and very different to the Thomas Organ ss VOX amps. But he makes some very useful suggestions.

[R.G.]

The amp has low power with only 1.4 volts at the speaker with four or eight ohms.

It's probably a bad idea to put a four ohm load on this amp if the power amp is like the T60. Too much current in the output transistors. Solid state amps have the reverse tolerance of a tube amp for loading. They work fine into an open circuit, but get the load too low and they can die. I wouldn't go under 8. On the other hand, it's fine with no speak on the output jack.

The voltages look fine.

That's the right first step. If the power supply isn't right on any amp, it won't work right.

I replaced the power amp transistors.

Was one or more actually bad by test, or did you do this as a precautionary measure?

Is tracing the signal the same on a ss amp? Input signal to a transistor a certain strength and higher at the output?

No. Sometimes it is, but transistor circuits also sometimes work on current gain more than voltage gain, so signal voltage does not always go up. In fact, this is one of those amps where it happens.

The output is a stacked, or so called 'totem pole' emitter follower. The driver transformer converts a roughly 20V swing on the collector of Q5 to a 5-6V swing on the secondaries, but at higher current.

It's a peculiar power amp circuit, with some real quirks. First, there is no DC feedbck - well, or AC feedback, either! - from the output, so the DC balance of the output stage is totally dependent on the matching of both transistor gain and biasing of the two output devices. Those OC28s are germanium power devices, which were always rare and expensive, and today even more so. I hope you didn't go buy a matched pair. If so, it had to hurt your wallet. You can convert to silicon PNPs, but it's harder with this version of the stacked circuit than with the Thomas Vox ones.

From your description, you may have a bad driver transformer. I hope not, because the only cure is to un-wind and re-wind that one if you want to keep the same cirucit. There aren't any replacements available. This is based on the comment that voltages are OK, but low signal out. You need to test the driver transformer by unhooking the primary and both secondaries, applying a test signal and seeing if the test signal makes it through. If the transformer is hard to drive with a signal oscillator at something like 1kHz, then it may have an internal short. You can test for opens and shorts between windings with your DMM, and use the neon-bulb internal short tester from geofex. A bad driver transformer is very rare, but your description does fit the symptoms.

Count on replacing any original electrolytic capacitors. Just do it. That *might* fix what ails it. If it were mine, I'd remelt all the solder joints too, shotgun style. Once you have fresh solder joints and fresh electros, start trasing signal and testing the driver transformer.

What signal strenght should the signal be at the input to the power amp transistors?

It's a few volts on the secondary winding at full power. Maybe 4-7V. It's more a current drive than a voltage drive. The primary at Q5 collector drives it to full power with a 15-20V peak signal. The driver transformer is single ended, so the primary sits with only a volt or two of resistance losses across it. Q5 collector pulls its end of the primary down on one polarity and lets it discharge above the power supply to the primary on the opposite polarity for full swing.

I get different readings at the Collectors of the power amp transistors. Is that normal?

Yes. the upper and lower sides of the output are both emitter followers, but it's a very odd arrangement. One collector stays at the negative supply, the other rides up and down with the output. It's how this stage works.

Can you help? Where can I go to learn more???

This will sound fishy, but I've actually spent about 20 years chasing down how that circuit works. It was only used for a very short time, and all the old textbooks I've pored over only at best mentioned it briefly. Most of what I learned was from circuit simulation and measurements on actual amps.

[AmpRX]

The amp has low power with only 1.4 volts at the speaker with four or eight ohms.
It's probably a bad idea to put a four ohm load on this amp if the power amp is like the T60. Too much current in the output transistors. Solid state amps have the reverse tolerance of a tube amp for loading. They work fine into an open circuit, but get the load too low and they can die. I wouldn't go under 8. On the other hand, it's fine with no speak on the output jack.

I should have known. The schematic even shows the acceptable range. The amp has three 4 ohm (probably really 3.5) in serie. It reads 8 ohms.

The voltages look fine.
That's the right first step. If the power supply isn't right on any amp, it won't work right.

Maybe not. Some of the voltage readings that are not on the schematic seems strange to me but I don't know what is correct. The readins on the power transistors are"

C: -37
B: 0.2
E: 0.004

C: 0.006
B: 37
E: 37

On the power supply after the 200 ohm resistor I read -25 volts.

The readings on the around TR5 are:

C: -22
B: -0.8
E: -0.6

There are some differences in this circuit to the one available on the web. If you have the schematics from the VOX Guidebook this is a comglomeration of those three. The biggest changes are C17, C18 and R34 deleted. C16 and R33 are deleted and the E of TR4 goes to gnd.

I replaced the power amp transistors.
Was one or more actually bad by test, or did you do this as a precautionary measure?

They were working fine but I changed one to see if there was any change. None. I put the original back. Eventually it shorted so I changed both. Today the other shorted. Obviously I don't know what I'm doing! I thought I installed them correctly.



It's a peculiar power amp circuit, with some real quirks. First, there is no DC feedbck - well, or AC feedback, either! - from the output, so the DC balance of the output stage is totally dependent on the matching of both transistor gain and biasing of the two output devices. Those OC28s are germanium power devices, which were always rare and expensive, and today even more so. I hope you didn't go buy a matched pair. If so, it had to hurt your wallet. You can convert to silicon PNPs, but it's harder with this version of the stacked circuit than with the Thomas Vox ones.

I bought two CV7085 from Langrex in the UK. They are supposed to be direct replacements and they seemed to work the same.

From your description, you may have a bad driver transformer. I hope not, because the only cure is to un-wind and re-wind that one if you want to keep the same cirucit. There aren't any replacements available. This is based on the comment that voltages are OK, but low signal out. You need to test the driver transformer by unhooking the primary and both secondaries, applying a test signal and seeing if the test signal makes it through. If the transformer is hard to drive with a signal oscillator at something like 1kHz, then it may have an internal short. You can test for opens and shorts between windings with your DMM, and use the neon-bulb internal short tester from geofex. A bad driver transformer is very rare, but your description does fit the symptoms.

Primary: 46 ohms
Both secondaries: 4.2 ohms

With a 1 volt 1k signal in, both outputs read .29 volts.

Count on replacing any original electrolytic capacitors. Just do it. That *might* fix what ails it. If it were mine, I'd remelt all the solder joints too, shotgun style. Once you have fresh solder joints and fresh electros, start trasing signal and testing the driver transformer.

All electrolytics even in the signal path were changed. I did that first. I had once worked on a JMI Defiant and it needed that to be done.

What signal strenght should the signal be at the input to the power amp transistors?
It's a few volts on the secondary winding at full power. Maybe 4-7V. It's more a current drive than a voltage drive. The primary at Q5 collector drives it to full power with a 15-20V peak signal. The driver transformer is single ended, so the primary sits with only a volt or two of resistance losses across it. Q5 collector pulls its end of the primary down on one polarity and lets it discharge above the power supply to the primary on the opposite polarity for full swing.

one side of the secondary reads 4.5v, the other side 1v. The 1v side is the one with the two lower voltages on the base and emitter.


This will sound fishy, but I've actually spent about 20 years chasing down how that circuit works. It was only used for a very short time, and all the old textbooks I've pored over only at best mentioned it briefly. Most of what I learned was from circuit simulation and measurements on actual amps.

Not fishy at all. Thanks for all the help. You are the man!!!!!!! I've been on your site before and the info was very helpful. YOu have done a great service to the community!

Mitch

[R.G.]

The readins on the power transistors are"
C: -37
B: 0.2
E: 0.004

C: 0.006
B: 37
E: 37

Probably OK. If you were curious, you could meter just between base and emitter. I suspect they'd both be about 200mV.

On the power supply after the 200 ohm resistor I read -25 volts.
The readings on the around TR5 are:
C: -22
B: -0.8
E: -0.6

Something is wrong, or at least inconsistent, here. The schemo shows 16.5V for that - and at the other end of the wire toward C9, also 9.8V. Those can't both be right - especially since you measure -22 on Q5's collector. What that tells me is that unless the circuits are substantially different, something is messed up with Q4/Q5 (I notice I'm subbing in "Q" for "TR"). If the schemo is even close to correct, there is not enough standing/idle current there to properly drive one polarity of the output stage.

Here's how that works. The driver transformer sits at idle with very little voltage across it, only the idle current of TR5 through the resistance of the copper in the primary. When TR5 increases its current because of an increasing voltage/current on its base, its collector pulls down on the end of the primary attached to its collector.

That pull down on the primary makes a transformed-ratio voltage appear on the secondaries. These are set up so they are reversed in polarity on the top and bottom transistors. So pulling down on the end of the primary turns on one output transistor and turns the other off. With no feedback, which is which is kind of arbitrary, but **they must be opposite** so only one output transistor turns on at a time.

[ It occurs to me that this is a fruitful area for you to check; try the signal-into-primary again, but this time figure out which wire of the secondaries are in phase and check to be sure that the + end of one drives the base of one output, but the - end of the other secondary drives the other output device.]

So with the signal going up at the base of TR5, the signal on the secondaries goes up on one output device base and down on the other. The output base that's pushed up by its secondary conducts more, and the output voltage goes in the direction that this output wants to move it. TR6 pulls the output negative, TR7 pulls the output positive. But one of them, determined by the phasing of the two secondaries pulls the output its favored direction.

When the signal on TR5 base reverses, TR5 tries to conduct LESS current than it does at idle. The primary, acting like an inductor, reverses its voltage and pushes the collector of TR5 beyond the power supply voltage at C9. This also reverses the polarity on the secondaries, so that the output transistor that was turned on with a positive going signal on TR5 base is turned off, and the opposite one is turned on. The two secondaries and the base-emitter diodes of the output transistors form a phase splitter, turning on one or the other output transistor at a time.

The current in the primary of the driver transformer **will** continue to flow, so whatever TR5 does not eat is forced into the base-emitter of the output transistor. And this forms one limit on this output stage. The current into the "turning off" output transistor cannot be any larger than the primary current starts at, and will in general be limited to the idle current of the primary times the Nsec/Npri ratio. So one output direction on these amps can drive as hard as TR5 can pull it by pulling down on the primary. The other side can only go on as hard as the idle current it gets by TR5 turning off.

And all that means the idle current in TR5 is critical. Your measurements show it to be lower than the schematic, even the least current-version of the schematic. It would be helpful to know the voltage from end-to-end of R20, 21, and 22 (using the T60 schemo numbers).

... and the E of TR4 goes to gnd.

I think you mean TR3. TR4 emitter can't go to ground and have the power amp work enough to pass audio.

You measure 0.6V at emitter of TR5; if R21 is 10 ohms as per the schematic, then the idle current is 0.6/10 = 0.06A. The schemo says by the same calculation that this should be 0.09A. Accordingly, the voltage across R22 should be 200*0.06 = 12V for the current you show or 200*0.09 = 18V for the schemo conditions. The schemo doesn't show the raw voltage into the end of R22, so I can't calculate what they think should be there.

Your measurement of the raw voltage on R22 is -37V on one side and -22V on the collector of TR5. That gives 15V drop in R22 and the primary of the transformer. The resistance between -37 and TR5 collector should be 246 ohms. 15 volts divided by 246 ohms is 61mA, corroborating the current in R21, as it should. So your circuit is running on only 2/3 the bias current the schemo alludes to.

I changed one to see if there was any change. None. I put the original back. Eventually it shorted so I changed both. Today the other shorted. Obviously I don't know what I'm doing! I thought I installed them correctly.

Be sure to test that they're insulated from ground on the heat sink when you re-mount them. And do worry about the phasing of the secondaries as I mentioned.

Primary: 46 ohms
Both secondaries: 4.2 ohms
With a 1 volt 1k signal in, both outputs read .29 volts.

You might try that with a bigger signal on the primary, or feed 1V into one secondary and measure again, just because accuracy suffers at low voltages. But the good news is that a bad driver transformer probably would not give those readings, so we'll consider it good until proven bad. I'm more on TR4 and TR5 now.

All electrolytics even in the signal path were changed. I did that first. I had once worked on a JMI Defiant and it needed that to be done.

Good step!

one side of the secondary reads 4.5v, the other side 1v. The 1v side is the one with the two lower voltages on the base and emitter.

Is this measured on an AC range on a meter, or with an oscilloscope?

It's kind of consistent with the limited idle current theory. One side is limply driven. I'm glad this is pointing away from the driver transformer - other stuff can be replaced.

[AmpRX]

[QUOTE=R.G.;220232]Probably OK. If you were curious, you could meter just between base and emitter. I suspect they'd both be about 200mV.

120mv


And all that means the idle current in TR5 is critical. Your measurements show it to be lower than the schematic, even the least current-version of the schematic. It would be helpful to know the voltage from end-to-end of R20, 21, and 22 (using the T60 schemo numbers).

R20: 793mv
R21: 628mv
R22: 12.7v
R23: 35v
Secondary connected to R23: 140mv
R27: 36.3
Secondary connected to R27: 145mv

I think you mean TR3. TR4 emitter can't go to ground and have the power amp work enough to pass audio.

Correct!

Is this measured on an AC range on a meter, or with an oscilloscope?

Meter

Here's the kicker. Someone did shoddy work on the amp before I got it. Wires were connected incorectly and there was very little output. I reconnected according to the schemo and the signal looked wrong on the scope. I continued to do some work and then posted my plea on this forum for help. I had connected the amp to my load and scope using the red wire from the speaker out to the hot wire and the black to the gnd. Well, today I noticed that the amp's red wire was gnd and the black was the speaker out. Bad to assume! I reversed polarity and voila, 21v with an eight ohm load! Sorry but this has been a good lesson in early ss circuitry. Thanks for all the help.

[AmpRX]

One more thing. There is some hum and hiss. I'm assuming (not a good thing as I found out) that some of that is normal since this is a very primitive ss amp. Any thoughts on whether or not the slight hum and hiss is normal? Thanks!

[R.G.]

[QUOTE=AmpRX;220326]

Wires were connected incorectly ... I reversed polarity and voila, 21v with an eight ohm load! Sorry but this has been a good lesson in early ss circuitry. Thanks for all the help.

Good catch. It wasn't going to work with the output miswired! Glad you caught that, and congratulations.

One more thing. There is some hum and hiss. I'm assuming (not a good thing as I found out) that some of that is normal since this is a very primitive ss amp. Any thoughts on whether or not the slight hum and hiss is normal?

Some hiss is unavoidable. At the time these were designed, the techniques for designing for low noise were not well understood, and early germanium devices were poorly packaged, so the hiss per device was worse.

I can make some suggestions if you want to tinker with noise performance. A real go-for-broke rework would be to replace TR1..TR3 with low noise silicon PNPs, and replace resistors R1..R5 with metal film. The circuit looks like it may bias up with silicon because of the way it's hooked up. This has to be considered experimental, though.

The hum is something else, though. The amp was badly reworked before you got it. It is likely that the rewiring has made the hum worse.

[AmpRX]

Today I jumped in an additional 1000 uf to each main cap that is supposed to be 2500uf (the replacements I found were 2200uf). It lowered the hum on one side. It could be that one of the new caps is faulty or it really needs 2500uf. The ones I put in are very small and rated at only 50 volts. I'm waiting for a couple of higher rated caps that I ordered the other day.

[R.G.]

You can get hum from trivial misplacement of the transformer center tap along the ground wire to the circuits. The current pulses from rectifiers are so high that they can induce hum when the ground to the preamp is merely placed in the wrong spot along the grounding. The transformer center tap must go to the first filter cap, no where else. You can also get current-induced hum/buzz from an input wire lying along the transformer and rectifier leads.

Get out your scope and look at the hum frequency. If it's 60Hz, it's power line pickup. If it's 120, it's ripple from the power supply or pickup of rectifier buzz.