Hello,
I have one of these now on my bench for repair. They have a unique power supply stage with SCR's in place of bridge diodes. The SCR's are supposed to behave like in a halogen light dimmer, stabilizing the voltage. Unfortunately, instead of +-70V, I get +-84V which is wrong, it seems, as the main caps are rated for 75V DC. The amp is also buzzing rather loudly (low-frequency spikes) on both channels. On the other hand, offset is perfect both L and R, so it seems the power mosfets and the preceding transistors are ok.
Can anyone please explain to me how this kind of regulation works, and what to check as it seems all four SCR's are full on, even in idle ?
Thank you in advance
Marco
I have one of these now on my bench for repair. They have a unique power supply stage with SCR's in place of bridge diodes. The SCR's are supposed to behave like in a halogen light dimmer, stabilizing the voltage. Unfortunately, instead of +-70V, I get +-84V which is wrong, it seems, as the main caps are rated for 75V DC. The amp is also buzzing rather loudly (low-frequency spikes) on both channels. On the other hand, offset is perfect both L and R, so it seems the power mosfets and the preceding transistors are ok.
Can anyone please explain to me how this kind of regulation works, and what to check as it seems all four SCR's are full on, even in idle ?
Thank you in advance
Marco
For those wishing to help Orit here is a better ( clearer ) breakdown of its constituent circuit parts , it does have its faults like injecting noise into the mains due to the operation of the SCR,s and the constantly variable voltage rail can induce clipping which is indicated on the power amp casing .
Its obvious that the control section of the power supply is faulty and having smoothing capacitors whose working voltage is said by Orit to be 75V DC when the power supply runs at 70V DC is not something I would do but obviously they have great faith in such a small window of control ----that is until a fault develops and that's where this unique circuit has its flaws .
When the fault is found I would change the overvoltaged capacitors , I also notice an SCR optocoupler is used .
You could test the SCR,s I built a simple tester for both SCR,s and Triacs but its probably faulty in (as I say ) the control section.-
Soundcraftsmen PCR800 Power Amplifier Technical Review – AmpsLab
Its obvious that the control section of the power supply is faulty and having smoothing capacitors whose working voltage is said by Orit to be 75V DC when the power supply runs at 70V DC is not something I would do but obviously they have great faith in such a small window of control ----that is until a fault develops and that's where this unique circuit has its flaws .
When the fault is found I would change the overvoltaged capacitors , I also notice an SCR optocoupler is used .
You could test the SCR,s I built a simple tester for both SCR,s and Triacs but its probably faulty in (as I say ) the control section.-
Soundcraftsmen PCR800 Power Amplifier Technical Review – AmpsLab
I absolutely don't get what these SCR's actually do in the main power supply. As I don't see any closed loop, I suspect there's also no supply voltager regulation. But what else for might the SCR's have been preferred over common diodes?
Best regards!
Best regards!
They vary the voltage rails depending on the current used to make for a cooler amplifier/smaller amplifier but that has its flaws and it also has a safety feature that lowers the wattage down to about 40Watts if overheated .
It was never popular although its pretty unique .
Thats another feature , while safety conscious I am not keen on due to its action on the overall fidelity of the amplifier
It was never popular although its pretty unique .
Thats another feature , while safety conscious I am not keen on due to its action on the overall fidelity of the amplifier
+ -15V power supply for thyristor control, fan optothyristor, protection OPamp.
Thyristors in a full-wave rectification circuit with a capacitive load stabilize (lower) the output voltage.
Adjust the power supply on a load equivalent - powerful resistances or incandescent lamps included instead of power amplifiers.
Topic thyristor power supplies and automation unit.
Thyristors in a full-wave rectification circuit with a capacitive load stabilize (lower) the output voltage.
Adjust the power supply on a load equivalent - powerful resistances or incandescent lamps included instead of power amplifiers.
Topic thyristor power supplies and automation unit.
Thank you dear friends for your answers.
I didn't mention that R33 in this circuit, a 10 ohm resistor that connects the rail ground with the chassis, is burnt. But I don't think that replacing it would change things.
I didn't mention that R33 in this circuit, a 10 ohm resistor that connects the rail ground with the chassis, is burnt. But I don't think that replacing it would change things.
Make sure that the 10 ohm resistor is replaced with the same value and wattage as its connected to part of the low voltage power supply and it controls part of the 15V+/- supply .
A peculiar concept. Around that time, the Yamaha A-760/960 amplifiers used a primary-side thyristor job for regulation.
The OP will have to test every single component related to the PCR - there aren''t that many anyway.
The OP will have to test every single component related to the PCR - there aren''t that many anyway.
Looking at the schematic SCR's CR20-23 , the gates for positive appear to me to have ability to turn on being biased to do so, however a capacitor is needed at the diode junction of CR18/19 to ground,as presently R25 may be coupling ripple
The negative though is a vipers nest of issues, I would hazard at R29 and R32 being
instead of facing the incoming AC supply, be arranged to access the anodes of the SCR's
It though may prove a chicken before the egg issue, that the circuit needs some referencing to its incoming AC supply before the SCR turns on. If so deriving gate triggers from small slave bridge rectifiers I know works well with thyristors, as they can be teased with partial DC.
I've made some checks last night to the control circuit on the upper left of the power supply schematic.
U2, like U1, is a quadruple op amp. U2B in particular, gets pulses from pin 6 and should have a DC voltage on pin 5, is supposed to deliver pulses to the four SCR's. This op amp has now a fully negative output, no pulses.
U2C and U2D seem to be part of the circuit that senses B+ and B- because the first is referenced to ground, and the second to a portion of a zener voltage, through a buffer transistor. Outputs from these op amps is fully negative, making pin 5 of U2B fully negative as well (C2 electrolytic is working with the wrong polarity !) so that output pin 4 is as described.
The +/- 15V voltage stabilizers are working correctly. U2D puzzles me more than the rest: emitter voltage of the buffer is 5.1V, but pin 14 is 1.6V - compared to 5V of pin 13 - because R10 and R13 make a voltage divider and output pin 12 is fully negative. Could it be that changing position to the cursor of R8 potentiometer fixes things up ?
U2, like U1, is a quadruple op amp. U2B in particular, gets pulses from pin 6 and should have a DC voltage on pin 5, is supposed to deliver pulses to the four SCR's. This op amp has now a fully negative output, no pulses.
U2C and U2D seem to be part of the circuit that senses B+ and B- because the first is referenced to ground, and the second to a portion of a zener voltage, through a buffer transistor. Outputs from these op amps is fully negative, making pin 5 of U2B fully negative as well (C2 electrolytic is working with the wrong polarity !) so that output pin 4 is as described.
The +/- 15V voltage stabilizers are working correctly. U2D puzzles me more than the rest: emitter voltage of the buffer is 5.1V, but pin 14 is 1.6V - compared to 5V of pin 13 - because R10 and R13 make a voltage divider and output pin 12 is fully negative. Could it be that changing position to the cursor of R8 potentiometer fixes things up ?
Thanks for pointing this out and opening my mind! Now I got it how the supply is being regulated. Those +70V and -70V flags indeed are somewhat hidden between the opamps.
Best regards!
I can understand how SCRs regulate voltage there.
Since once triggered they stay ON until voltage/current cross zero, and triggering them early would allow voltage to charge capacitors charge to peak voltage anyway, only way for them to have any influence is to trigger *after* peak was reached, on the down slope.
As a byproduct creating a brutal charge pulse which no doubt must be coupled to mains, making "everything else" buzz, which seems to have been a common complaint way back then.
And ripple waveform must have been ´orrible.
Do not pretend to understand that phase control and the fuzzy scan does not help much.
To boot, it´s incomplete, we don´t even see the d*mn power transformer!!!!
Nor its connections!!!!
Guess SCRs are fine (failure mode would be one or more shorted and lots of smoke or at least *exploding* fuses), but they are being turned ON all the time, for all 180 degrees of mains waveform (or almost).
Try to restore control board to original working order, you know I hate shotgunning but in desperate cases .....
You see something burnt? ... replace it.
Being a dyed in the wool minimalist and thousands of kilometers away from good suppliers, **I** would add a bucking voltage transformer knocking down mains voltage by 15% so raw, unregulated rails become +/-70V instead of current +/-84V and it behaves like any other honest to God unregulated supply amplifier. (99.999% of amps out there)😎
But that´s me, huh?
MAYBE some transformer primary taps in some combination allow for the voltage reduction I suggest as a shortcut.
IF you have its schematic, show it.
Since once triggered they stay ON until voltage/current cross zero, and triggering them early would allow voltage to charge capacitors charge to peak voltage anyway, only way for them to have any influence is to trigger *after* peak was reached, on the down slope.
As a byproduct creating a brutal charge pulse which no doubt must be coupled to mains, making "everything else" buzz, which seems to have been a common complaint way back then.
And ripple waveform must have been ´orrible.
Do not pretend to understand that phase control and the fuzzy scan does not help much.
To boot, it´s incomplete, we don´t even see the d*mn power transformer!!!!
Nor its connections!!!!
Guess SCRs are fine (failure mode would be one or more shorted and lots of smoke or at least *exploding* fuses), but they are being turned ON all the time, for all 180 degrees of mains waveform (or almost).
Try to restore control board to original working order, you know I hate shotgunning but in desperate cases .....
You see something burnt? ... replace it.
Being a dyed in the wool minimalist and thousands of kilometers away from good suppliers, **I** would add a bucking voltage transformer knocking down mains voltage by 15% so raw, unregulated rails become +/-70V instead of current +/-84V and it behaves like any other honest to God unregulated supply amplifier. (99.999% of amps out there)😎
But that´s me, huh?
MAYBE some transformer primary taps in some combination allow for the voltage reduction I suggest as a shortcut.
IF you have its schematic, show it.
Thanks for all your suggestions my friends. JMFahey the windings of the mains transformer are (badly) shown on the bottom of the schematics. I too think that the SCR's are not faulty, if just for statistics: all 4 with the same fault would be strange.
Yes turning this kind of Vincent motorcycle to a normal power amp is a possible way out ...
Yes turning this kind of Vincent motorcycle to a normal power amp is a possible way out ...
I also don't think the SCR's are toast. If they were, they were either open or, rather, short circuit. In both cases you'd see no DC voltage at all.
Best regards!
Best regards!
From the description:№2
What is so special about this amplifier is the PCR part, which is an acronym for Phase Control Regulation. To understand this term better, it means the power supply is Regulated. And this regulation is made possible by controlling the phase angle of the thyristors feeding the filter capacitors, hence the term Phase Control Regulation.
The transformer secondaries are connected at the left. They are at 68-0-68Vac. After rectification, the output is at +/- 70Vdc. This voltage is regulated, meaning it doesn’t sag under load. I can verify this because when I bench tested this PCR800 at full power into an 8Ω dummy load, the supply rail voltages stayed at +/- 70Vdc.
The red box at the top is the phase control section. This is the heart of the PCR. This circuit controls the firing of the thyristors. Since the voltage is regulated at +/- 70V, it indicates that the thyristors are fired before 90°. There is a +/- 25V headroom in the power supply because in a conventional, non-regulated linear supply, the output will be at +/- 95V.
The circuit enclosed in blue is for temperature sensing.
What is so special about this amplifier is the PCR part, which is an acronym for Phase Control Regulation. To understand this term better, it means the power supply is Regulated. And this regulation is made possible by controlling the phase angle of the thyristors feeding the filter capacitors, hence the term Phase Control Regulation.
The transformer secondaries are connected at the left. They are at 68-0-68Vac. After rectification, the output is at +/- 70Vdc. This voltage is regulated, meaning it doesn’t sag under load. I can verify this because when I bench tested this PCR800 at full power into an 8Ω dummy load, the supply rail voltages stayed at +/- 70Vdc.
The red box at the top is the phase control section. This is the heart of the PCR. This circuit controls the firing of the thyristors. Since the voltage is regulated at +/- 70V, it indicates that the thyristors are fired before 90°. There is a +/- 25V headroom in the power supply because in a conventional, non-regulated linear supply, the output will be at +/- 95V.
The circuit enclosed in blue is for temperature sensing.
Rectified voltage is +/- 96V under conditions stated.
OP measured +/-84V only because transformer is being heavily overloaded, caps must be leaking like crazy (dangerously heating up in the process), power amp must be pulling huge idle current, etc.
All dangerous and unsustainable, it´s a miracle it´s not popping fuse caps, smoking resistors and overheating or damaging semiconductors, etc.
True, but that needs a working controller.
Best solution is to repair it, it might be as easy as replacing the burnt resistor mentioned, who knows?
IF NOT, "safe" Plan B would be to reduce secondary voltage to around 50-0-50 VAC
The transformer is big and expensive, so I suggested adding a bucking voltage one in series with primary.
Please measure actual VAC present at main secondaries today, to calculate ne auxiliary transformer,which won´t be too large or expensive.
Just hope there is free space inside the chassis to add it.
It´s actually *after* the 90 degree point, when sinewave is already going down.
If it were before, capacitors would charge to peak anyway; this is a capacitive input supply, not a lamp filament or universal motor load.
Again, best is to try to repair the controller.
Plan C would be to design a small circuit board, triggering SCRs (using Opto Diacs or SCRs?) at the proper point after zero crossing, say 100/110/120 degrees or 6-7 milliseconds too late (for 50Hz mains) so rectified voltage is at +/-70V
You could adjust that with a trimmer pot by adjusting triggering delay.
Basically what the current circuit does , only non-regulating.
Not sure temperature control would work, you might want to add a couple bimetallic heat sensors to heatsink for safety, something similar to fan motor or plain let it run 100% of the time, like many do.
Main problem here is that it´s both a complex amp BUT think perceived value is not too high ("I can buy a similar one on EBay for xxxx$") so I doubt customer will want to spend much on repairs.
Plan D would be to replace current PT with one salvaged from a dead similar amp.
Tons of amps out there with around +/- 70V rails so around 50+50VAC secondaries.
NOTE: remember Fan motor is 120VAC only, currently connected across *one* of dual 120V AC primaries.
Last edited:
I have removed one SCR for testing. At my 20€ component tester, it looks like a 45 ohm resistor between the gate and the cathode. Anode disconnected. Tried with 20V and a 1kOhm resistor in series, but I haven't been able to fire it, while I nearly fired a 10 Ohm resistor from +20V to the gate trying to switch it on.
If what you wrote there is true, then the op-amp IC chip is bad and that's why it's allowing high voltage. If pin 13 is higher than pin 14; then the output should be positive. Seems like it is shorted negative. And that will pull pin 5 of U2B low and trigger the SCRs on as soon as they can be. That will overcharge those caps and put a load on the transformer while it cooks them.
And you are correct, R8 is for setting the voltage. Should set it to 140V between positive and negative supplies.
Last edited:
Thyristors turn on from a gate current supplied from the anode side, so R29,32 1k ohm resistors actually bias it off until a desired current is met. Helps prevent false triggers.
The negative side is the best way to do that. It is a proper mirror image to the positive side of things. This is because the positive side is triggering from an AC supply to DC, and the negative side is triggering from DC to AC. So the diodes are still needed where they are, as are R29,32. Things on the DC side are consolidated to reduce parts, such as R27,31. The reason Q5 and Q6 are N and P channel is because the triggers at the bases need to be buffered from each other so they can't flow through and turn them selves on. And they also need to be triggered at the exact same time so there isn't a delay between rail firings to creat noise harmonics. This timing is accomplished by Q3,4. They turn on at the same time from the same source and provide buffered outputs.
The referencing of the AC signal comes from Q1. It discharges the phase circuit every half AC cycle, or 120hz when the cycle crosses 0V. When charging, voltage on C1 ramps up until it is reset. This voltage is seen as a rising saw-tooth wave on pin 6 of U2B. Voltage on pin 5 is the comparator control voltage. The lower it is, the earlier in the phase cycle that the SCRs are triggered. Higher voltage means later. This can go from full on to full off. The control voltage is determined by the voltage regulator of U2C,D and the temp protection from U1A,C,D.
Hope that clarifies things a bit.
Last edited: