How Audio Amplifier BIAS / Idle Current Circuit Works : A Real Life Example FP10000Q 10Kw Amp

追加: 2026-05-18

[00:00:00]Hi guys, welcome to learn electronics repair and the video I published about these amplifiers caused a lot of interest because we seem to have an impossible situation here. If we just looked at a few of the comments on that video and thank you everybody who did comment by the way some themes are coming up regularly and this is the video I was referring to.

[00:00:25]So uh Jim Hawksley works on jukebox amplifiers says that sometimes you have to set the quesscent current after changing the output transistors. Sometimes it's difficult because the trim pot is faulty and you put usually sorts it. Now I actually can adjust the quescent current or idle current. The pot is actually working. I just can't adjust it enough.

[00:00:51]Another one here fox junkie. So, he's describing a problem where he's repaired amplifiers, sent them back out, they blew up again, and it's because the customer had the wrong speakers or the wrong combination of speakers attached.

[00:01:06]These amplifiers come from a proauudio rental company and sales. They basically know what they doing, these guys.

[00:01:15]They're not electronics repair experts, but they do know what they are doing.

[00:01:18]And in fact, I've had amplifiers brought in when they've told me that a blown speaker blew the amplifier. So, they know how to identify that. I think we can probably discount it in this case, but it's a good point. Okay. Always ask the customer that. Doug, quite an in-depth discussion about how the bias circuit works. The dead king's raven wanting to know if I when I was repairing the PCB damage, I shorted the speaker output to ground. We can test that. Astro's electronics lab, he's agreeing with me about the idle current adjustment because it does actually adjust.

[00:01:56]Concentrates on measuring the voltage is at the voltage multiplier Q22. Okay.

[00:02:03]Compare with the working one. And if they're wildly different between that and the other one, then something wrong with the multiplier Q22 circuit. Okay, this one from Doug again and a very similar response from Carson Napia.

[00:02:24]So, wildly different currents on the MPN and PMP sides of your output amp. Where is that current going when you have no speaker attached? It's weird.

[00:02:34]Absolutely. That's one of the strange things we see here. And just to reply to John Burns, yes, you can power audio amplifiers up without speakers connected. There's no problem in doing so. In fact, it's the safest way to test them after repair before you connect the load. So, why do I know something weird is happening and some of you guys did?

[00:02:56]Well, to understand this, you have to understand a little bit about the circuit and about the bias on class AB amplifiers. What is it? Why do we need it? So the quick down and dirty discussion about bias with these amplifiers in general. We have a positive supply.

[00:03:18]In fact, we always have a positive supply and we have a negative supply and between the two we have our output transistors often with several of them in parallel.

[00:03:30]Okay.

[00:03:33]emitter resistor which is a low value typical 0.22 ohms.

[00:03:42]This is your speaker out.

[00:03:45]Okay. And then we have another emitter resistor, PMP transistor.

[00:03:56]And as I mentioned, this is your V minus base resistor.

[00:04:06]And this is where the audio feeds in.

[00:04:10]And with a class AB amplifier, this transistor handles the positive going part of the audio waveform and this handles the negative going. So, as you can see, if you increase the voltage on the base, this transistor will conduct.

[00:04:32]While I'm here, let's put this in. So, you can see it balances. Okay, this conducts and the speaker output becomes more positive. The higher the voltage of the waveform, the sine wave, the more this conducts, the more voltage appears at the speaker and the more the cone of the speaker moves. Okay? And then when the sine wave comes back down to zero and goes negative, the PMP transistor conducts because in this case the base needs to be negative with respect to this point. Okay, this goes more negative and the more negative it goes, the more the transistor conducts and the more it pulls the speaker voltage down to the negative rail. So that's how it works. The higher powered amplifier, the more voltage you need. So low powered amplifiers maybe 4550 volts.

[00:05:32]The high powered ones I work on, don't be surprised if it's 160 volt.

[00:05:38]And that's because for a given resistance and impedance of the speaker, to get more current flowing through it, i.e. more wattage, you need more voltage. Okay. The speaker attaches to ground.

[00:06:00]Yes, bridged amplifiers connect between the two, but keep it simple. Okay, speaker is going to ground. Now, the advantage of this circuit is that only one of these transistors is turned on at a time. Positive half the waveform, negative half. So when there's no input, the transistors are not on. So no current's flowing. Uh it's efficient. So it doesn't pass current when there's no audio.

[00:06:28]But there's a problem. The transistor, if you know, I'm sure you know a bit about transistors, will not start conducting until the base is about 0.6 volts more than the emitter. Or in the case of the PMP, the base is 0.6 6 volts less than the emitter. But with this audio waveform, any audio waveform, it passes through zero, it passes through zero. So when the voltage here is close to zero or close to zero in the negative, neither of the transistors are switching on. So in this area here, neither transistors are switched on. And if you didn't design a circuit to allow for that, what you'll end up with is a sine wave that looks something like this.

[00:07:33]Okay?

[00:07:35]Not drawn to scale. So this is where bias comes in. What we need to do is bias our transistors so that with no audio coming in they are just conducting they start to conduct. So to do that we need about 6 volt on the base relative to the emitter and we need about 6 of a volt on this base negative to the emitter and that's where the bias comes in.

[00:08:06]So the most simple method would be used for resistors. Of course we need to feed the audio signal in from here.

[00:08:42]capacitor, nonpolarized, by the way. It's just the way I drew it.

[00:08:48]And we feed the audio in here. So, because we're using the capacitor, we only apply AC here. We don't affect the DC voltage here, and we don't affect the bias. So, that is a basic bias circuit.

[00:09:03]Although you don't generally see that, especially not in high powered amplifiers, but that's the principle of it. Another very similar setup which I will mention because you're more likely to see this on low powered amplifiers.

[00:09:15]Instead of these resistors, we have two diodes in series like that. You know, the base of a transistor base emitter drops about 6 volt. Well, so does a diode. Yeah, silicon diode drops about the same. And the advantage of this arrangement is that as the transistor warms up the voltage drop between the base and emitter decreases. Okay. But the same is true of the diodes as well. So if you mount the diodes close to or on the heat sink where the transistors are, as the transistors warm up, the diodes also warm up and it counteracts for this change in the actual voltage drop base emitter. So that's why we see that sort of circuit in the real world. Of course, and you've probably noticed this, things are always more complicated.

[00:10:23]So this is the bias arrangement on the amplifiers I'm working on at the moment.

[00:10:28]You can see here these are the output devices. Three transistors in parallel MPN and PMP. Okay.

[00:10:38]This is the driver stage. So this transistor is driving the base. You can see there are no base resistors in there. Not from that point at least.

[00:10:48]Okay. Same on this one. The driver stage is driving the base. There are resistors in the circuit by the way. You can see here from the base we have resistor down to this is your speaker out, speaker out and there are emitter resistors on each of the transistors.

[00:11:08]And here is our bias network. So we have on each side like a totem pole. We have a transistor here. So connected via a resistor to the plus 160 volt that then connects to this transistor.

[00:11:25]This is our idle current or bias adjust here VR1.

[00:11:30]And then that connects down to an MPN transistor the complement of that one.

[00:11:37]And this goes down again via 82 ohms to the minus 160. So this forms a voltage divider using resistors and transistors. And this one sits in the middle. As you can see with this little network of resistors, two fixed, one variable, we can actually alter the voltage drop across here. And the voltage drop across here is going to be basically the difference between the two base voltages.

[00:12:08]So across the transistor about 1.2 2 vol. Or you might think there should be 1.2 volts across here, but you'd be wrong. The reason you probably think that is because in this explanation of how the bias circuit works, we have the two diodes in this case. So we have 1.2 volts between there and there. Well, this is single stage amplifier. Ours isn't because if we look at the collector circuit of the bias transistor and the emitter circuit will be the same on the PMP side. Here we can see via resistor it sets the bias on this transistor MJ340. This is the pre-driver. The emitter of this one drives the base of the next one. That is the driver.

[00:12:55]Okay. And then the emitter of that one drives the bases of all the output transistors. So we don't have one semiconductor junction drop 6 volts here. We actually have more because it's like a daisy chain. So basically the positive bias and the audio of course comes into the first transistor and the emitter of that one goes into the base of the next stage pre-drive it drive it and the emitter of that goes into the base as along we draw one the output transistor then this goes to speaker out and we have the same thing upside down on the PMP side. So to bias this on you have to have enough voltage to overcome all these three base emitter junctions. And we know they're about 6 voltage. So this is about 1.8 volt or if you like 3 times voltage base emitter.

[00:14:01]Yeah. And we'll have the same on the PMP. Another minus 1.8 volt for exactly the same reason. So the bias voltage is going to be somewhere in the region of 3.6 volts. So we should be able to measure that across the biasing transistor. So across here, yeah, we'd expect about 3.6 volts. And again, this transistor is on the heat sink, so it will compensate for the fact the drop varies with a rise of temperature.

[00:14:34]Okay, these two transistors, by the way, they're doing two things. One is to make this a constant current supply. So the current through the transistor is fixed, which means that this will not vary. You can set the offset voltage between the collector emitter. Don't have to worry about the current changing. And this is where we feed the audio signal in. You can see the audio is coming in here, okay, through these resistors into this circuit. This is a differential pair.

[00:15:05]And that differential pair is driving into this transistor you can see on this one.

[00:15:12]And they drive into the two transistors in the bias circuit which we just talked about. So not only are these transistors maintaining a constant current, they're also feeding the audio signal in to the driver and output transistors. Yeah.

[00:15:31]here and here. You can also see the base of the driver transistors is controlled by the voltage across this one. So that's how the bias is applied into the amplifier and how the audio signal gets in. Each of the output transistors has an emitter resistor. One, two, three.

[00:15:54]There you can see them. One, two, three.

[00:15:58]There you can also see them.

[00:16:00]So these are connected to the emitters of each of the output transistors.

[00:16:06]And this is speaker out. This red line.

[00:16:09]Okay, that is speaker out.

[00:16:13]If you wondered what this little what here is doing, by the way, that pair and this pair of transistors, you'll see that effectively the base of this one is driven by the voltage drop across the emitter resistors there. Okay? And the same on this side. So, effectively these are monitoring the voltage drop across the emitter resistors amplified through a couple transistors. plenty of gain and that also via this diode here and this diode here feeds back in to this bias network. Yeah, you can see that this for the positive half cycle, this for the negative half cycle. That's why the diodes are opposite way rounds, of course. So, why is this monitoring the voltage across the emitter resistors and then feeding something back into the base of the driver? Well, with biasing you have really a compromise. So the more you reduce the biasing to the point where the transistor cuts off and stops conducting, the more you can reduce the quiescent current, the idle current. But to get a linear amplification across the amplitude of the waveform, you need to have more bias current. That's because a transistor doesn't just turn on at 0.6 volts on the base.

[00:17:45]This is current voltage.

[00:17:52]Okay. The transistor passes no current.

[00:17:55]Doesn't switch on until we reach the forward bias voltage on the base.

[00:18:04]let's say 6. At that point, the transistor doesn't just like turn on. It starts to conduct in the curve sort of like this. And eventually, it reaches a point called saturation.

[00:18:17]Or at least it's called saturation now.

[00:18:22]Back in the 70s, it was called hard on.

[00:18:28]You had a hard on. Why they changed that? Well, down there. But that's how I was taught in the 70s. Anyway, apart from those semantics, you can see that at this point when it starts to switch the transistor on, the curve is not linear and as it approaches saturation, it's not linear. The linear part of the amplification isn't here. Okay? So, you want your waveform to always stay within this area of the response of the transistor. If it goes below here, then you'll effectively cut off the bottom part of the waveform.

[00:19:05]And if it goes above there, you will cut off the top part. So to get our transistor with a very low quiescent current, we would kind of set the bias to here.

[00:19:16]But to get it to actually amplify with much less distortion, we need to get to the bias to about here. And obviously that gives you more idle current. So that's what this circuit is doing and it's complement on the PMP side. It's monitoring the voltage across the emitter which basically tells us how much the transistor is conducting and it actively adjusts the bias depending on the amplitude of the waveform. So that's what that is and you'll see this in a lot of amplifiers by the way. So if you ever see that again you'll know what it's doing. That same circuit by the way is driving these opto isolators. You can see here. So when the LEDs are lit, the transistors conduct. One end goes to ground and the other end of that is driving the limit signal lights up the limiting LED. So what is so weird about our amplifier?

[00:20:18]Well, I have a good one, known good one, and I have the bad one. And if you look the good one, this is the current I see from my vent power supply. And I can adjust that with the idle current control. But this is not all the idle current through the output transistors because this has this tracking power supply class TD with book converters in.

[00:20:44]And even when there's no audio input, those book converters are working and they're drawing power. Okay, so this is normal for this style of amplifier. You can see 15 watts from the negative supply, 16 from the positive. The extra little bit of current is driving some other circuit on this board. It's not part of the amplifier processing the audio.

[00:21:07]And across the emitter resistors, I can see the voltages. And you can see they're all very similar. And from this we can actually work out the current passing through each transistor because each transistor has its own emitter resistor. And what we find is I equals V over R Ohm's law. Voltage is a voltage I am reading and R is a value of the resistor. And on my amplifier nominally the resistors are 0.1 ohms.

[00:21:43]And the voltage I'm seeing, let's take the average of this 0.12.

[00:21:49]And the voltage here, I've just written it down so I didn't get the notes mixed up. So that's one volt, 100 mill volts, 10 m volts, 1 mill volt. And ours is point something of a mill volts. One, two. And if we divide that by that, we get this. So the current is 1.2 milliamps.

[00:22:17]Remember there are three output transistors and three emitter resistors.

[00:22:22]So the same is flowing through each one.

[00:22:25]I measured each one. So the actual idle current is three times that which is 3.6 milliamps idle.

[00:22:39]And the supply voltage from the book converters plus or minus 6 is the same.

[00:22:44]That's good. What's happening on the bad one?

[00:22:48]Well, we can see on the PMP transistors, I got between 1.97 and 1.92. So, we'll call the average of that 1.95 molts. So, 1.95.

[00:23:04]So, we're getting a mill volts. 0.0 0 0 1 9 five quite a lot more. Yeah. Divided by 0.1 ohms is 19.5 milliamps.

[00:23:27]So a lot more idle current. Okay.

[00:23:31]And on the NPN transistors it's even worse. I read between 5.4 49 and 6.49 measuring the three resistors. That could suggest the resistors are not so equal in value, but it could just be a fact of more current shows more difference. Okay, so let's say the average of that is six. And once again, 6 milliamps divided by 0.1 is 60 milliamps. So, how do we know something strange is going on? And only a few of you picked up on this. This is the good amplifier. So, we can see the measurements I took. We know the value of the resistors.

[00:24:20]We know the voltage drop. There was slight differences.

[00:24:24]0.11, 0.12, 0.13, but we'll just say that's the average. Okay. So through each resistor we have 1.2 milliamps and there's three effectively in parallel.

[00:24:38]So we get 3.6 milliamps flowing from the positive supply to here.

[00:24:45]And if we look on the PMP side we see exactly the same thing. So same voltages same currents 3.6 milliamps flowing back out. And that's where the current is flowing. It's flowing from here to here.

[00:25:03]It's not flowing out there. Well, a because there should be no volts on here anyway if it's balanced, which is same voltage as ground. So, no current would flow. But even more importantly, we don't have a speaker attached. Yeah. So, it can't flow that way. But what's happening here? Well, on the MPN side, I got some wildly different readings, but the average is six. So we'll say 6 m volts and we use the same calculation with the 0.1 ohm resistors and we find there are 60 milliamps flowing through each one. You know that's a lot more total 180 milliamps. A lot lot more.

[00:25:45]And on the PMP side we had roughly an average of 1.95 millolts with some slight differences. So each resistor carries 19.5 milliamps and the total is 58.5 milliamps. Let's say 60 for easy maths. So there's 180 milliamps coming down here. There's 60 going down there.

[00:26:10]There's no speaker attached. Where's your other 120 milliamps going? Yeah, told you something strange happening here. And we can see this clearly on the schematic. This red line is the junction between all those resistors. This goes to a out audio out. That's your speaker.

[00:26:30]And there's no connection to ground.

[00:26:34]That's a capacitor. This is a triac. If this was switched on, then yes, we would have a connection to ground. Uh CG, probably circuit ground or something.

[00:26:46]Chattery ground more than likely. But there's a couple of reasons why I'm certain that isn't switching on.

[00:26:52]Firstly, because well, we changed it on the bad one and we tested with the curve tracer, but we will test it again with the multimeter and it's not shorted. And the other reason this actually is driven by a DIA and a DIA will switch on at 32 volts.

[00:27:14]It's kind of like a birectional 32 volt zen if you like. And we can't have 32 volts on here for the simple reason the supply rails via our book converters are plus and minus six.

[00:27:29]So I don't see how a that can be switched on and b how it's possibly conducting. But we can test. Do we have a short from here to here? Is that where the current's going? Just coming back this way. Well, there's nowhere current can really flow to ground. Nothing goes to ground. Okay, so that's a bit of a mystery if we really have what we can see we have. I messed some numbers off here, so we'll just put them on now. This was plus 6.49 vol and this was - 6.49 vol. So they're exactly balanced.

[00:28:11]whereas those aren't exactly but they're close.

[00:28:16]Why do I say if we really have what we are reading? Well, because this is a logical there are some possibilities. One is some straight path to ground which you just discussed about and the other one is we working out the current based on the fact that the resistor is the value it says it is. If those resistors are higher in value than they should be, and resistors do fail high, then the current calculation is wrong because we're dividing this by 0.1. And if you divide it by something more than 0.1, the answer gets smaller.

[00:28:56]Yeah. And the same here. So there's a problem with the resistors. In actual fact, we have the same current flowing through each path. Therefore, that paradox has kind of gone away in a puff of logic or is this really happening? Uh, I will just add I did measure the image resistors and it did measure. Okay, so we're going to have to take some readings here now to prove this. What is really happening? We'll start with the good one. So, these are the transistors we can see in that network which sets the bias. So 21 is one on the MPN side, 22 is the middle one and 12 is the one on the PMP side. So we can measure the voltage on there and we can see do we actually have the cross collector emitter here the voltage we've just calculated about 3.6 6 V and all the resistors we can see the same voltage drop in the microvolts range. They were all basically identical. But we can confirm this. If we lift one end of a resistor, we can actually I won't use the clamp meter. It's not sensitive enough. But I can put my meter multimeter into the circuit and we can see what actual current is flowing through that resistor. To do that, obviously, I'll have to tag a wire back onto the pad where I've lifted the end of the resistor from to insert into the circuit. I can do with either end.

[00:30:28]I'll just do one resistor on this because we know this one works anyway.

[00:30:33]On the bad one, we can do the same thing.

[00:30:38]So, the voltages around these transistors and these ones.

[00:30:43]And I'll lift especially on the MPN side because I'm getting quite different readings. I'm going to lift all three and we'll measure the current through all three. And then we'll measure the current through one of these as well and we'll see do we really have what we think we have. Another good indication I may be onto the right track here is that this is the current being drawn on the good one and this is the current on the bad one.

[00:31:11]And this is from the positive and the negative rails. And you can see there's about 10 milliamps more on the positive compared to the negative. But the same applies on both. The currents are higher, but the balance between the two is the same. Yeah. So that also very much indicates we don't have a lot of current going to ground somewhere from the positive rail. Otherwise, it would show there. Which also just raised another quick point in case you've wondered how can we have 180 milliamps here when we only have 210 milliamps there. The difference is 30. But we can see on the good one we have 134. Yeah, that's because this current is flowing from a 6V supply.

[00:31:59]That's not the same as having 60 milliamps on the 120 volt supply.

[00:32:05]In fact, a bit of mental arithmetic off the 120 volt rail, it would be about 9 milliamps. Yeah. But we'd still see it and we're ready to go.

[00:32:28]So, we can see the current there, the 10 milliamps different. We can see the actual current flowing through the resistors is not what we thought at all.

[00:32:40]6 of a milliamp and they're pretty well balanced. Okay, so that's on the emitter resistors on both sides.

[00:32:49]So measuring the voltage and calculating the current was very incorrect the result. So this is the PMP side. So this is the collector minus 1.8 and the emitter 120. Okay, just make a note of that value on the positive side. That's this one minus 1.8 1.8.

[00:33:23]Yeah. What does our calculations tell us?

[00:33:26]And then on the other one, plus 1.77 plus 120.

[00:33:38]And I didn't make a note of the idle current.

[00:33:44]You can see there it's actually climbed a little bit. I think it's because it's warmed up at the time for a little while. So, there's some warmth in this now, but it's still pretty low to be quite honest. That Let's say about 0.8 milliamps.

[00:34:02]And that feels really satisfying because we looked at this circuit. We looked at the way the bias is applied, the number of transistors in the output stage.

[00:34:16]Okay. And we calculated we should see about 3.6 volts across here. And we do in fact exactly what we expected. Yeah, I measured on both transistors, but they connect to each other. In actual fact, I didn't need to. So, we can see plus 1.8 volt minus 1.8 volts. These 220 ohm resistors feed into the pre-driver.

[00:34:42]And that is the actual current. So, it's not like we were calculating from measuring the resistors, but I didn't use a milliohm. Yeah. Let's do the same on the bad one. And here we are.

[00:35:01]So yeah, about 10 milliamps difference there.

[00:35:05]What do we see? We see -4.6.

[00:35:10]Oh, so it is very different. -4.6 milliamps and plus 7. So it was telling the truth. What's on these transistors?

[00:35:23]1.9 minus 1.3 and on the Well, it's going to be the same.

[00:35:34]Yeah. Minus 1.3 120 + 120 1.88.

[00:35:46]Now, and this is very telling. So we can see this is actually 1.9 volt positive not 1.8. So those transistors are turned on more than they should be on the MPN side. Hence the fact we have 14 milliamps and not8.

[00:36:06]Okay. And on this side we have minus 1.3 volt.

[00:36:13]So they are turned off but we still see about 7 milliamps. So, there's still current flowing down through there and now we really need to find out why. But we certainly know pretty much what's going on. I still can't just quite figure out where that extra current actually flows to. And bear in mind that the total current flowing is three times that. Okay. So, about 45 milliamps.

[00:36:38]A lot less than I actually thought it was. Okay. And the current flowing through here is actually quite a bit less than I thought it was as well.

[00:36:48]Okay, I'm going to order the transistors then guys and we can continue with this.

[00:36:52]But we need to make sure we have all the parts because it may be we have to repair this by substitution.

[00:36:58]We can certainly see what's going wrong in the bias network. So, I think we can look at this first, get the bias in correct, and then see what we have there because this is definitely wrong. And it may be wrong either because there's a problem with these 82 ohm resistors or there's some problem with one of these transistors or more than one. Yeah, but just before we finish, let's just check something else. And there's something else just to measure these emitter resistors. This is the good one. But I just have one lifted on this that reads.20.

[00:37:35]Let me just check the zeroing on my meter.

[00:37:41]06.

[00:37:43]And if I measure the cross resistor again, yeah.15 or 16. So it's 0.1. I'm sure this side will read the same because this amplifier is balanced properly. Okay.

[00:37:55]Yeah. So, let's just look on the other one to see if this is part of the problem. And if we don't see a difference, I'm going to use the millio meter.

[00:38:08]I think that tells us enough. Oh, no.

[00:38:10]It's reading down to that. Yeah. Oh, it was basically the same.

[00:38:20]In fact, I'd say we don't even need to use the mill. That's good enough.

[00:38:26]So emitter resistors are okay.

[00:38:30]Output devices on this one are all taken from that one and they were working on that one. So what's left probably is this. What next? Uh and maybe some 82 ohm resistors.

[00:38:46]R48 is tucked right down here.

[00:38:50]81.6. This is the bad one. And R53 is here.

[00:38:57]Yeah.

[00:38:59]And let's just be certain we do not have a short from ground to speaker out.

[00:39:12]I'd say there's a capacitor charging just messing up the auto ranging there. Let's switch auto ranging off.

[00:39:24]60 megga ohm range. Okay. So ground to speaker out there. It's about 300k and the other way around it's about 100k something charging there. You can see how does that compare with the other one.

[00:39:49]Couple hundredk or something. Same thing with the capacitor charging again. Capacitor charging 100k. Yeah, about the same really. So, I'm basically back where I was on the previous video. MJ 340s, MJ 350s, and this one D1691Y.

[00:40:12]I could start stripping down another good working one, but I've got a suspicion the problem isn't here. And anyway, you guys will want to talk about it. Yeah. So, I'll get those parts ordered. You guys get chatting down there. And I look forward to seeing you all soon again on Learning Electronics Repair. Ciao for now, guys.