I have a cheap 5.1-ch 80W computer stereo, that I'm completely rebuilding, and on all new prototype board (long story). I don't have a schematic's, so I've just copied all values, from the original PCB.
I've had it 95% rebuilt and working fine, powering it from a bench PSU, with the main Vcc as 17V, Vee is just GND, and the op-amp common-mode/reference rail is 6V. The original had a transformer for the main Vcc, and that was fed into what seemed to be a SMD 7806 regulator.
Short version is, running 6V rail off PSU works fine, but using a 7806, it's voltage is all over the place, the DC current is meant to be under 1mA....because NONE of the op-amps IP/OP, has a DC path to GND...so maybe it's a minimum output current issue for the 7806?
the basic schematic picture is below, I only added the 1 op-amp, just to symbolize it. And the the terminal regulator should be the 7806
TLDR
It's meant to run off a transformer for 17V, and from that into 100R, into what seemed to be, a generic SMD 7806 , and into a 10R resistor, for the 6V rail, which is like the common-mode voltage for the op-amps. With lots of filter caps.
There's no DC path to GND in any of the RC networks of the op-amps, so when it's working, the current draw on the 6V rail , is well under 1mA. And all the inputs and outputs of the op-amps are basically 6VDC, with a little bit of offset voltages/currents.
Then today as soon as I soon as try to power it all off the 17V rail, and with a new TO-220 7806, it can't keep 6V, it's jumping up and down by volts on the output of the 7806, and the op-amp OP's are copying it, and/or driving it, but it's basically 1:1 and in phase on the scope.
I have copied all the input/output caps and resistors that were from the transformer to the op-amps, and 7806. Nothing else uses 6V.
And again I stress, there's no DC path to GND, on any iputs, or output's, of this pre-amp section, and it's the only thing that uses the 7806.
AND I un-soldered the 7806, and hooked 6V rail back to the bench PSU, and the stereo works again, the 6V rail is pulling under 1mA, and all op-amps are at 6V like they should be.
So fearing a short on my proto-board, I put the 7806 on a breadboard with the the 100R and 10R, and caps too, and it worked for a while, then now, it powered it back up, and guess what the 7806 output voltage was, 0.9V, and I'd imagine the op-amp voltages where all over the place.
And I had this exact problem on the old PCB, too, but figured it was something else, or a short(they where there too). And I had this problem with new SMD 7806, and it seemed there was a short circuit.
So what's going on?
I've had it 95% rebuilt and working fine, powering it from a bench PSU, with the main Vcc as 17V, Vee is just GND, and the op-amp common-mode/reference rail is 6V. The original had a transformer for the main Vcc, and that was fed into what seemed to be a SMD 7806 regulator.
Short version is, running 6V rail off PSU works fine, but using a 7806, it's voltage is all over the place, the DC current is meant to be under 1mA....because NONE of the op-amps IP/OP, has a DC path to GND...so maybe it's a minimum output current issue for the 7806?
the basic schematic picture is below, I only added the 1 op-amp, just to symbolize it. And the the terminal regulator should be the 7806
TLDR
It's meant to run off a transformer for 17V, and from that into 100R, into what seemed to be, a generic SMD 7806 , and into a 10R resistor, for the 6V rail, which is like the common-mode voltage for the op-amps. With lots of filter caps.
There's no DC path to GND in any of the RC networks of the op-amps, so when it's working, the current draw on the 6V rail , is well under 1mA. And all the inputs and outputs of the op-amps are basically 6VDC, with a little bit of offset voltages/currents.
Then today as soon as I soon as try to power it all off the 17V rail, and with a new TO-220 7806, it can't keep 6V, it's jumping up and down by volts on the output of the 7806, and the op-amp OP's are copying it, and/or driving it, but it's basically 1:1 and in phase on the scope.
I have copied all the input/output caps and resistors that were from the transformer to the op-amps, and 7806. Nothing else uses 6V.
And again I stress, there's no DC path to GND, on any iputs, or output's, of this pre-amp section, and it's the only thing that uses the 7806.
AND I un-soldered the 7806, and hooked 6V rail back to the bench PSU, and the stereo works again, the 6V rail is pulling under 1mA, and all op-amps are at 6V like they should be.
So fearing a short on my proto-board, I put the 7806 on a breadboard with the the 100R and 10R, and caps too, and it worked for a while, then now, it powered it back up, and guess what the 7806 output voltage was, 0.9V, and I'd imagine the op-amp voltages where all over the place.
And I had this exact problem on the old PCB, too, but figured it was something else, or a short(they where there too). And I had this problem with new SMD 7806, and it seemed there was a short circuit.
So what's going on?
Attachments
Last edited:
A standard linear regulator like a 7806 can only source current, it cannot sink it, and if you try to make it, the voltage simply rises immediately. The dynamic currents due to the signal could be a lot more than 1mA, and you need to ensure there is a bias current to ground from the output of the 7806 so that it always has DC load current. Try 100 ohms 1W resistor load or such, that might be all you need to fix this.
[ assuming my understanding of your circuit is correct ]
[ assuming my understanding of your circuit is correct ]
I never did find the SMD code on the original, it was like E6 or something, but it was wired exactly like any other 78xx, so it wasn't like a LM317 adjustable regulator, can they sink a certain amount of current like a TL431 ? I could totally use a TL431 tho I guess.
The only thing missing from the main Vcc rail is the 6800uF, maybe that charging slower helped the original start up balancing act?
Ok thanks I'll add a resistor, but only a few mA, I hope it doesn't shag the AC of the opamps.
I'm still fairly new to electronics
The only thing missing from the main Vcc rail is the 6800uF, maybe that charging slower helped the original start up balancing act?
Ok thanks I'll add a resistor, but only a few mA, I hope it doesn't shag the AC of the opamps.
I'm still fairly new to electronics
Last edited:
A few mA is unlikely to be enough - I suspect the original device was a shunt regulator, not a series regulator
Your reasoning is astray. Each opamp has a path from Vcc to ground via its PSU pins, and 12 opamps drawing a total of 1mA is simply not credible. It would be more like 50mA if not a great deal more.
pull the output of the regulator to gnd with a 470 to 2.2k resistor to give it some load
TL431 set to 6v with a 2.2k to 17v would work.
TL431 set to 6v with a 2.2k to 17v would work.
Op amp output voltages are proportional to the supply voltage....
Also check that the 7806 is able to take 17 volts input, if needed just add a series resistor on the input side.. Or change to a 7812 if the pre amp chip allows it.
You may need to make some minor changes for that if needed.
And add a 220 pF or so to filter the SMPS noise, which was not there in the transformer based supply.
Also check that the 7806 is able to take 17 volts input, if needed just add a series resistor on the input side.. Or change to a 7812 if the pre amp chip allows it.
You may need to make some minor changes for that if needed.
And add a 220 pF or so to filter the SMPS noise, which was not there in the transformer based supply.
I've been mostly improvising my board layout as I go, and I left off a few things from the power rails.
So tidied up all that, and really had it like on the original PCB, this week, but still powering it off the PSU, not the transformer, or it's rectifier and 6800uF cap.
But each pair of dual op-amps, has it own 100R and 47uF/100nF, on it's Vcc. And on the 6V rail, there's about 4-5 x47uF, and also 3x470uF caps laid around.
And added a 1.5k resistor to the regulator's output, for 4mA like the data sheet said, and I added a protection diode on the OP pointing at the IP.................and it still doesn't work, the regulator's OP went to 0.9V.
So I'll try again w/ the transformer andmain filter cap, maybe that will be the right start-up charge up time.
I said that last time but I had to fix up the rest of the power rail, and hoped it would have done it.
Other than that it's works OK when it's working, but I still have a funny problem that I'm not sure if I damaged the rear input volume chip, or if I'm getting some feedback loop. It hasn't happened so far today after I cleaned the board a bit more.
It's in the fine tuning stages now anyways, I'm almost done.
So tidied up all that, and really had it like on the original PCB, this week, but still powering it off the PSU, not the transformer, or it's rectifier and 6800uF cap.
But each pair of dual op-amps, has it own 100R and 47uF/100nF, on it's Vcc. And on the 6V rail, there's about 4-5 x47uF, and also 3x470uF caps laid around.
And added a 1.5k resistor to the regulator's output, for 4mA like the data sheet said, and I added a protection diode on the OP pointing at the IP.................and it still doesn't work, the regulator's OP went to 0.9V.
So I'll try again w/ the transformer andmain filter cap, maybe that will be the right start-up charge up time.
I said that last time but I had to fix up the rest of the power rail, and hoped it would have done it.
Other than that it's works OK when it's working, but I still have a funny problem that I'm not sure if I damaged the rear input volume chip, or if I'm getting some feedback loop. It hasn't happened so far today after I cleaned the board a bit more.
It's in the fine tuning stages now anyways, I'm almost done.
They most certainly are not. They are proportional to the difference between the input voltages and nothing else, or to the single input voltage in a feedback arrangement, until the output level hits the rail voltage or thereabouts, when it clips.
I breadboarded some of the pre-amp's and all the output chipamps, but notthe whole thing together. So now with it soldered up it's sort of a pain to fix stuff.
1 thing I should try also, is with the 7806 on a breadboard, take some single-shots of the start up, see what's actually happening. I can current limit my PSU so overall nothing gets too toasty.
1 thing I should try also, is with the 7806 on a breadboard, take some single-shots of the start up, see what's actually happening. I can current limit my PSU so overall nothing gets too toasty.