But that's why we adjust the pot... 🙂
It's likely that aside from this last issue, you've got it working.
Sealed pots are good. The ones that give trouble are the cheap ones that lay flat on the board, wide open and filled with dust.
Okay, get your meter on either R44 or R45, start up the amp and begin adjusting it right away. Set the Bias to about 25 to 30 mv as I explained and get it as close to 25 as you can. Check it every couple of minutes and re-adjust as necessary. The time to worry is when we can't get that voltage down to reason --or-- the DC offset starts to drift on us, so you need to check R47 to ground after each adjustment as well.
The reason the voltage was climbing was that things were getting warm... the reason things were getting warmer is that the voltage was climbing. It's a very nasty trait of BJT outputs called "Thermal Runaway" and it can kill a perfectly good amplifier.
Let me know how you make out...
How much should the voltage be reduced? I suppose that you're talking about the main voltage 120VAC?
I do have a step up transformer 120V to 220V. I could use it in reverse mode 120V to 65V. Does that count?
BTW, there are no resistors R220 0.5W on the actual board or even on the layout that is on top of the schematic diagram. That's the funny part. There also mistakes in the pinout of each transistor (B-C-E are not where they should be), as well as mixing up their numbers (that's on the real board I'm working on)
I do have a step up transformer 120V to 220V. I could use it in reverse mode 120V to 65V. Does that count?
BTW, there are no resistors R220 0.5W on the actual board or even on the layout that is on top of the schematic diagram. That's the funny part. There also mistakes in the pinout of each transistor (B-C-E are not where they should be), as well as mixing up their numbers (that's on the real board I'm working on)
Attachments
Don't worry about it then ... just do the adjustment as I detailed. Lets see what that gets us...
Ok. According to you instructions, I got the volts across R44/R45 between 24.7 and 26.3 mV. When I go above or below that number, it in- or decreases, but much slower than the last time. Transistors are mildly warm to the touch. Ground and R47 reading is 7 mV on 1 side and 30 mV on the other.
Would it be safe now to connect a speaker to the output or is there something else that should be checked?
Would it be safe now to connect a speaker to the output or is there something else that should be checked?
I would say to let it burn in for a couple of hours, make sure those readings don't drift upwards or things don't get hot.
Sidenote: It doesn't take much to get those transistors warm, does it?
Once you are satisfied that it's stable, you can hook up a speaker and try some music tests. But, I would strongly suggest you start at an extremely low volume... so low that you have to put your ear up to the speaker to hear anything. You will be listening for distortion. If there is note worthy distortion at those low levels you may want to reset the bias to 30mv and repeat the monitoring process... I would not recommend you go beyond 30mv, though.
Don't go straight from there to full blast... it's better to play music at moderate "conversational" levels for a couple of hours while you keep an eye on the temperatures ... then move forward as you gain confidence in each step.
Sidenote: It doesn't take much to get those transistors warm, does it?
Once you are satisfied that it's stable, you can hook up a speaker and try some music tests. But, I would strongly suggest you start at an extremely low volume... so low that you have to put your ear up to the speaker to hear anything. You will be listening for distortion. If there is note worthy distortion at those low levels you may want to reset the bias to 30mv and repeat the monitoring process... I would not recommend you go beyond 30mv, though.
Don't go straight from there to full blast... it's better to play music at moderate "conversational" levels for a couple of hours while you keep an eye on the temperatures ... then move forward as you gain confidence in each step.
I turned the amp of in order to move it somewhere else. When I turned it on again the relay wouldn't kick in. On the anode of D16 there are 3.42 mV, on kathode are 3.51 mV present. The diode seems to be ok, although I got a reading when measuring reverse polarity but this could be the windings from the relay. It measured the same last time when the relay still activated.
For the bias adjustment...
Try turning it back down to 25 ... this could go on for quite some time before you get it stable. If you can get it to start at 20mv, maybe give that a try... It might settle at 25 for us.
Also keep an eye on the DC offset. If it gets over about 600mv it will trigger protection and prevent the relay from pulling in.
I've had a few amps where I've had to go around the adjust and wait loop half a dozen times. That we have no spec for what it should be doesn't help...
A small modification...
If you have a 4.7K resistor and an LED we can add a visual protect indicator quite easily ... See the thumbnail... The LED will glow when the amp is either in startup delay or protect mode and go off when the relay is pulled in. There's no harm in leaving this in the circuit for future troubleshooting.
The readings in your last post are a bit of a concern. If you aren't reading a steady 12v to ground on the cathode of D16 there may be a problem with the +12v power supply. If the anode switches from +12 to almost ground but the relay doesn't pull in, there may be a problem with the relay itself.
This might be a good time to check the solder and foil in that area of the board to be sure you have solid connections.
Also the schematic shows a switch marked S1 ... if that switch is turned on, the amp will stay in protection mode forever.
Try turning it back down to 25 ... this could go on for quite some time before you get it stable. If you can get it to start at 20mv, maybe give that a try... It might settle at 25 for us.
Also keep an eye on the DC offset. If it gets over about 600mv it will trigger protection and prevent the relay from pulling in.
I've had a few amps where I've had to go around the adjust and wait loop half a dozen times. That we have no spec for what it should be doesn't help...
A small modification...
If you have a 4.7K resistor and an LED we can add a visual protect indicator quite easily ... See the thumbnail... The LED will glow when the amp is either in startup delay or protect mode and go off when the relay is pulled in. There's no harm in leaving this in the circuit for future troubleshooting.
The readings in your last post are a bit of a concern. If you aren't reading a steady 12v to ground on the cathode of D16 there may be a problem with the +12v power supply. If the anode switches from +12 to almost ground but the relay doesn't pull in, there may be a problem with the relay itself.
This might be a good time to check the solder and foil in that area of the board to be sure you have solid connections.
Also the schematic shows a switch marked S1 ... if that switch is turned on, the amp will stay in protection mode forever.
Attachments
Last edited:
I got it running!
The relay possibly did not kick in 'cause I disconnected the thermal switch that is mounted on the heatsink, right above the tiny screw on the trimpot. And I forgot to reconnect it......
I got the amp stable at 26 mV with all volumes down, nothing connected. When I put a speaker to it, that bias rises when turning up the vol and decreases when turning it down. I was not aware of that.
I had it running with an old tone generator at 700 Hz for about 45 minutes, master vol at 4 o'clock and everything runs nice.
Now I can give it back to the owner.
Thank you so much for your help, Douglas.
Now I con
The relay possibly did not kick in 'cause I disconnected the thermal switch that is mounted on the heatsink, right above the tiny screw on the trimpot. And I forgot to reconnect it......
I got the amp stable at 26 mV with all volumes down, nothing connected. When I put a speaker to it, that bias rises when turning up the vol and decreases when turning it down. I was not aware of that.
I had it running with an old tone generator at 700 Hz for about 45 minutes, master vol at 4 o'clock and everything runs nice.
Now I can give it back to the owner.
Thank you so much for your help, Douglas.
Now I con
Idle bias on emitter resistors can only be measured when the amp is silent. When music flows through emitter resistors, the voltage on them increases. Some meters may ignore the AC signal and only show the DC. Others will not.
Last edited:
Hi
Wipe of the old grease and DO NOT apply new !!
Thermal grease is supposed to fill MICRO voids in the surfaces and when aplied correctly you cannot see it. It is universally applied wrong so the best way for most people and companies to use it is to not use it.
As far as repairing solid-state PAs go, you almost always have to do it brute force - replace drivers and outputs in one shot, otherwise parts keep getting destroyed, especially the way you are testing without any power limiting.
Build the standard Power Limiting Safety Socket. This uses incandescent bulbs as power limiters, wired in series with one side of a receptacle. use a low-watt bulb for initial testing just to make sure there are no shorts and such like. You can do basic tests up to swept signals into a bench load, then change to larger lamps and gain confidence that the circuit is likely working.
You should be testing without loads first, checking DC voltages then using a sine wave test signal and scope to see what the circuit is doing. Note that even with a defective driver, say, most PAs will pass a clean sine wave without load - adding a light load of 50R will show up the defective part - half the wave may suddenly disappear.
Only after you've done thorough testing with a bench load should you ever connect a speaker. The DC detection and speaker relay may not respond quickly enough to save your speaker from having a whole DC rail applied across it - then it's a dead speaker. I did that testing my very first amp decades ago and have since used a very incremental (pussy?) test method and have not blown up parts or speakers beyond what began as blown up.
Wipe of the old grease and DO NOT apply new !!
Thermal grease is supposed to fill MICRO voids in the surfaces and when aplied correctly you cannot see it. It is universally applied wrong so the best way for most people and companies to use it is to not use it.
As far as repairing solid-state PAs go, you almost always have to do it brute force - replace drivers and outputs in one shot, otherwise parts keep getting destroyed, especially the way you are testing without any power limiting.
Build the standard Power Limiting Safety Socket. This uses incandescent bulbs as power limiters, wired in series with one side of a receptacle. use a low-watt bulb for initial testing just to make sure there are no shorts and such like. You can do basic tests up to swept signals into a bench load, then change to larger lamps and gain confidence that the circuit is likely working.
You should be testing without loads first, checking DC voltages then using a sine wave test signal and scope to see what the circuit is doing. Note that even with a defective driver, say, most PAs will pass a clean sine wave without load - adding a light load of 50R will show up the defective part - half the wave may suddenly disappear.
Only after you've done thorough testing with a bench load should you ever connect a speaker. The DC detection and speaker relay may not respond quickly enough to save your speaker from having a whole DC rail applied across it - then it's a dead speaker. I did that testing my very first amp decades ago and have since used a very incremental (pussy?) test method and have not blown up parts or speakers beyond what began as blown up.