Hmm, you mean the output stage base drive signals? I'll ignore protection circuits for now. So you're saying that if the output stage gets driven from the front end, VCC/VSS drops to 2V (with series 100R resistors in the rails).
Try this:
VCC --> 100R ---> Output Stage --> 100R --> Input/VAS.
That is, power up the amp from a single power supply. Use 100R for the rail fuses. Input/VAS gets the same power too but from a RC circuit (aka original Leach).
Measure voltages across C2, C15, C22, C28, C19. I'd like to see these voltages without any input/output. Please post them here.
If the basic DC operating conditions are met, we'll figure out what next to check.
sounds like your bias is set to turn the back end on fully - check the voltage that the front end provides to the driver bases. It should be less than 2V
Terry, I have found the protection transistors in the VAS stage to be particularly susceptible to getting smoked, at least the tiny TO-92 packaged ones that Leach uses. These only switch on when things are out of wack, but when they do, they can't seem to handle the wack. These are T7 and T15 on Jens' boards.
Thanks guys,
Arius,
I'm not sure I can power it up with both connected without smoking the resistors but I'll try. When the voltage output from the filter caps is 30V the resistors are dropping 28V. I'm not sure how long they can do that.
If I can get it to hold I'll post the voltages you asked for. The weird thing is, that connecting it with the regulator on the front and the Filter caps feeding the back end, as I bring up the voltage with the Variac, everthing looks ok until it reaches a certain voltage and then it's like something turns on and the drop over the safety resistors starts to rise sharply. The voltage on the back end begins to drophe higher I raise the voltage with the variac, the lower the voltage on the back end falls.
Hi Bob,
I'll check that again. I checked it with the back end unplugged and I believe it was around .5V. I remember that the voltage for the base of T9 was +1.5V and the base of T13 was -.5V. On Jens' chart it shows 1.7v for both, so I know I have something wrong. he doesn't show a voltage for the base of T11 and Ithink something may have been wrong there. I got different readings for TH1 1&2 on my good board than I did on my bad one. The last thing I did today was replace the diodes D6, 8, 9 & 11 but I didn't have a chance to test it before I had to leave for work.
Hi Pooge,
I replaced most of the transistors but I'm not sure if those two were part of it. I would think that with rail voltage on the output that the protection circuit would try to kick in.
I'll replace those as well.
thanks so much to all of you for the help.
Blessings, Terry
Arius,
I'm not sure I can power it up with both connected without smoking the resistors but I'll try. When the voltage output from the filter caps is 30V the resistors are dropping 28V. I'm not sure how long they can do that.
If I can get it to hold I'll post the voltages you asked for. The weird thing is, that connecting it with the regulator on the front and the Filter caps feeding the back end, as I bring up the voltage with the Variac, everthing looks ok until it reaches a certain voltage and then it's like something turns on and the drop over the safety resistors starts to rise sharply. The voltage on the back end begins to drophe higher I raise the voltage with the variac, the lower the voltage on the back end falls.
Hi Bob,
I'll check that again. I checked it with the back end unplugged and I believe it was around .5V. I remember that the voltage for the base of T9 was +1.5V and the base of T13 was -.5V. On Jens' chart it shows 1.7v for both, so I know I have something wrong. he doesn't show a voltage for the base of T11 and Ithink something may have been wrong there. I got different readings for TH1 1&2 on my good board than I did on my bad one. The last thing I did today was replace the diodes D6, 8, 9 & 11 but I didn't have a chance to test it before I had to leave for work.
Hi Pooge,
I replaced most of the transistors but I'm not sure if those two were part of it. I would think that with rail voltage on the output that the protection circuit would try to kick in.
I'll replace those as well.
thanks so much to all of you for the help.
Blessings, Terry
I feel sorry for Terry.
What we (or I) want to do is to measure operating voltages at various point when the amp is at quiescent. To do that....
1) Both front and back ends have to be powered.
2) Rail voltages must exceed 40V for the input stage to work properly.
3) Bias pot, P1 cranked to max resistance. (Good point, Bob)
Under these conditions (and at optimum bias), Jen's voltage values are valid.
If something goes wrong when the back end is hooked up to the front end, then let's try to run on a minimalist config.
1) Like Pooge said, protection ckt may be toast. Remove T10, T12. Diodes D5, D14 may be suspect so with T10, T12 removed, protection ckt is out of the picture. Remove also D10, D12.
2) If step #1 doesn't bring DC conditions to normal, remove R37, R38, R64, R65. That is, run with one output pair. Fewer parts, fewer things that can go wrong.
Rail voltage at the output can only mean the output transistors are not driven right or they are shorted. Since you're testing without a load, the protection ckt won't kick in. It senses current and voltage. Please check also for mechanical faults (open/shorts) and resistor values with the amp powered off. Any circuit diagnostic we can do assume no faults of those type.
What we (or I) want to do is to measure operating voltages at various point when the amp is at quiescent. To do that....
1) Both front and back ends have to be powered.
2) Rail voltages must exceed 40V for the input stage to work properly.
3) Bias pot, P1 cranked to max resistance. (Good point, Bob)
Under these conditions (and at optimum bias), Jen's voltage values are valid.
If something goes wrong when the back end is hooked up to the front end, then let's try to run on a minimalist config.
1) Like Pooge said, protection ckt may be toast. Remove T10, T12. Diodes D5, D14 may be suspect so with T10, T12 removed, protection ckt is out of the picture. Remove also D10, D12.
2) If step #1 doesn't bring DC conditions to normal, remove R37, R38, R64, R65. That is, run with one output pair. Fewer parts, fewer things that can go wrong.
Rail voltage at the output can only mean the output transistors are not driven right or they are shorted. Since you're testing without a load, the protection ckt won't kick in. It senses current and voltage. Please check also for mechanical faults (open/shorts) and resistor values with the amp powered off. Any circuit diagnostic we can do assume no faults of those type.
Hi Terry -
Those 100R resistors would be 1/4 watt as they are meant to smoke if anything is wrong.
Those 100R resistors would be 1/4 watt as they are meant to smoke if anything is wrong.
Thanks Arius,
I will try what you suggest. As soon as I can get it done I will report back.
Hi Byrd.
I tried using 1/4 watt resistors. They smoked. I changed to 5w and use a light bulb in series with the AC line. So far I haven't smoked anything else. I don't think the 1/4 watters would have survived the 27V drop. I'm thinking that the 100ohm resistors and the light bulb are taking most of the current when there is a problem but I could be wrong about that.
Blessings, Terry
I will try what you suggest. As soon as I can get it done I will report back.
Hi Byrd.
I tried using 1/4 watt resistors. They smoked. I changed to 5w and use a light bulb in series with the AC line. So far I haven't smoked anything else. I don't think the 1/4 watters would have survived the 27V drop. I'm thinking that the 100ohm resistors and the light bulb are taking most of the current when there is a problem but I could be wrong about that.
Blessings, Terry
still4given said:
If you have P1 set for maximum resistance, (you checked this with your ohm meter, right?) the emitter to collector voltage of T11 will be less than the 3.4V (+/- 1.7V) shown in Jens' schematic. The base of T11 ought to be .6V above its emitter.
It sounds like there is an offset in the front end, If you adjust P1 does the voltage across T11 change? does it stay offset? Does the voltage at T11's emitter change?
You might want to pull the output transistors (I know 🙄 ) and check the board the way Prof. Leach suggests - you'll put a temporary resistor in to close the feedback loop. This should allow you to bring up all of the voltages long enough to troubleshoot.
The -.5 V at the emitter of T11 sounds suspiciously like a diode drop. Is the base of T11 somehow shorted to ground? What is the voltage at the collector of T6/R21? How do the input section voltages compare to Jens'?
put those 1/4 W resistors back - yes your light bulb helps you but the idea is to blow the little resistors if there is a problem - they act as current limiting fuses and help prevent further damage to the board.
In post 1100 you said there is 2 mV over your 100R to the front end. Was this a typo? - it should be closer to 2V. If not a typo, something is wrong in the front end.
Hi Guys,
OK, I'll try to answer in order.
OK, I disconnected the regulator and put 100R, 5W resistors in for R1 and R23. I also had 100R 5W resistors between the PSU and VCC, VSS. I powered it up using the light bulb in series with the AC just for my piece of mind. The voltage on the filter caps would only come up to 44 VDC and the rails would only reach 21.6 VDC behind the 100R resistors, no matter how high I turned the variac. There was as you can see almost 23V drop across the safety resistors.
I'm not sure if this will mean anything since I couldn't get the voltage any higher than 21 V but here goes.
C2, C15 20.9V each
C22, C28 21.1V each
C19 20.8V
1.2V, -1.2V
OK, I removed them and checked them with my meter. They seemed OK but I replaced them with new just in case. No change.
1) Check, as noted above
2) Couldn't get it to 40v with both front and back powered up.
3) I'm assuming that bias pot cranked means lowest ohm reading which was 00.3ohm.
OK I removed t10 and T12 as well as D5, D14, D10 and D12. I just replaced those zeners yesterday with matched units.🙁
Now when I try to bring up the power the rails will only go to .5V. Everything else gets burned off by the light bulb and the 100r resistors. I haven’t tried removing R 37,38, 64 and 65 yet. This is bumming me out the amount of parts I have replaced and things seem to be getting worse. I don't have time now to pull anything else and test before I leave for work. Will try tomorrow morning.
Hi Bob this is a little out of order because I checked your stuff first before I started removing the parts today but your post was last so I'm answering it now.
These voltages are with only the front-end energized through the regulator.
Rails -66.7, +66.2
I have P1 set at 00.3 ohm, T11 emitter = 2.72V. with P1 maxed in the other direction T11 emitter = 2.696
t11 b= -.9V e = -1.124V
As you can see I was wrong about the -.5V .
T6 collector = 64.8
Without the backend powered it reads 1.4V drop across the 100R resistor to the front end.
Ok guys, I've really got things screwed up now. I hope this info will help you help me.
Thanks, Terry
OK, I'll try to answer in order.
Yes
OK, I disconnected the regulator and put 100R, 5W resistors in for R1 and R23. I also had 100R 5W resistors between the PSU and VCC, VSS. I powered it up using the light bulb in series with the AC just for my piece of mind. The voltage on the filter caps would only come up to 44 VDC and the rails would only reach 21.6 VDC behind the 100R resistors, no matter how high I turned the variac. There was as you can see almost 23V drop across the safety resistors.
I'm not sure if this will mean anything since I couldn't get the voltage any higher than 21 V but here goes.
C2, C15 20.9V each
C22, C28 21.1V each
C19 20.8V
1.2V, -1.2V
OK, I removed them and checked them with my meter. They seemed OK but I replaced them with new just in case. No change.
1) Check, as noted above
2) Couldn't get it to 40v with both front and back powered up.
3) I'm assuming that bias pot cranked means lowest ohm reading which was 00.3ohm.
OK I removed t10 and T12 as well as D5, D14, D10 and D12. I just replaced those zeners yesterday with matched units.🙁
Now when I try to bring up the power the rails will only go to .5V. Everything else gets burned off by the light bulb and the 100r resistors. I haven’t tried removing R 37,38, 64 and 65 yet. This is bumming me out the amount of parts I have replaced and things seem to be getting worse. I don't have time now to pull anything else and test before I leave for work. Will try tomorrow morning.
Hi Bob this is a little out of order because I checked your stuff first before I started removing the parts today but your post was last so I'm answering it now.
These voltages are with only the front-end energized through the regulator.
Rails -66.7, +66.2
I have P1 set at 00.3 ohm, T11 emitter = 2.72V. with P1 maxed in the other direction T11 emitter = 2.696
t11 b= -.9V e = -1.124V
As you can see I was wrong about the -.5V .
T6 collector = 64.8
Without the backend powered it reads 1.4V drop across the 100R resistor to the front end.
Ok guys, I've really got things screwed up now. I hope this info will help you help me.
Thanks, Terry
still4given said:
A little confusing since you say two different voltages on T11 emitter - but if you really meant collector and have 2.96V that will cause your positive side of the output stage to turn on fully, and effectively short to the rail. Find that cause.
The emitter and collector should be the same absolute value - like you said early on +/-1.2 - so much is changing it is hard to keep up.
minimum bias is at maximum P1 resistance - keep it there until you sort out the rest to protect your outputs.
T11 emitter to base voltage is low. Have you swapped out this transistor?
Only change one thing at a time - it will make trouble shooting much easier.
I'm sorry to hear that you haven't fixed it yet. Can I suggest you bring the Krell back in from the garage for a week or so and take a break from the Leach. Sometimes, trying too hard works against you. I've been there. When you jump back in, sometimes something you have ignored suddenly pops up.
I'm sorry, I was answering your question about emitter to collector voltage. That measured 2.72V. The -1.24V was emitter to ground.
Yes I have replaced it. When I got bad readings there I replaced it hoping it would fix it but no change.
It'e really running weird now that I pulled the zeners and those two transistors T10 and T12.
I'm going to put new ones in there and start over.
Actually, what I think I'll do is pull the seft channel out of the case so I can get good measurements and make a voltage chart on the schematic and then compare that to what Jens posted. Then I'm going to go through the right channel board and compare and list the differences here.
Maybe somene will see a red flag and be able to tell where things are going south.
Well I might do that if I can't get it going within the next couple of days. My bench is covered with parts from this amp right now and I might not get everything back where it goes if I move them.
Blessings, Terry
Hi Terry
Just to be absolutly CLEAR ( have nearly done this myself a few times). The bias pot must be at MAXIMUM resitance. Your reading "00.3ohm" seems to indicate that it set to minimum resistance.
If this is the case it is quite possible that the output transistors have been blown. - Check them for shorts.
Just to be absolutly CLEAR ( have nearly done this myself a few times). The bias pot must be at MAXIMUM resitance. Your reading "00.3ohm" seems to indicate that it set to minimum resistance.
If this is the case it is quite possible that the output transistors have been blown. - Check them for shorts.
Well, I may be wrong but I always thought that a 10 ohm resistor had MORE resistance than a 100 ohm resistor. I know for sure that a 2 ohm speaker has more resistance than an 8 ohm speaker.
The bias go up as I turn the pot clockwise on the channel that works. If I turn the pot all the way clockwise it reads about 00.3 ohms. I cannot adjust the bias the way the bad channel is workings. There is no voltage drop across the output resistors. I doubt the outputs are blown. I just replaced all of them. They still read the same with an ohm meter as they did before I powered them up. The first thing I do almost every time before I power up is go through the whole board and check for shorts. If I have any shorts, they are not between any of the transistor leads nor the transistor leads to ground. It's always the first thing I check for.
Blessings, Terry
The bias go up as I turn the pot clockwise on the channel that works. If I turn the pot all the way clockwise it reads about 00.3 ohms. I cannot adjust the bias the way the bad channel is workings. There is no voltage drop across the output resistors. I doubt the outputs are blown. I just replaced all of them. They still read the same with an ohm meter as they did before I powered them up. The first thing I do almost every time before I power up is go through the whole board and check for shorts. If I have any shorts, they are not between any of the transistor leads nor the transistor leads to ground. It's always the first thing I check for.
Blessings, Terry
Good on the no shorting
I was also fooled into thinking about resistance max and min in the same way as you are viewing it. Always thinking about speaker impedance. However just as a 100uF capacitor has more capacitance than a 10uF capacitor a 100R resistor has more resitance than a 10R resistor. If you have a 0 Ohm reading you have no resistance, if your meter display error or -1 or something of that sort - it means infinite resistance - ie maximum resistance.
You want the bias current to be as low as possible and hence need to set the pot as such. In your case this appears to be all the way anti-clockwise.
I was also fooled into thinking about resistance max and min in the same way as you are viewing it. Always thinking about speaker impedance. However just as a 100uF capacitor has more capacitance than a 10uF capacitor a 100R resistor has more resitance than a 10R resistor. If you have a 0 Ohm reading you have no resistance, if your meter display error or -1 or something of that sort - it means infinite resistance - ie maximum resistance.
You want the bias current to be as low as possible and hence need to set the pot as such. In your case this appears to be all the way anti-clockwise.
Byrd said:
Ok, I understand what you are saying now. If you had said lowest bias rather than maxed out resistance I would have known what to do.
I will try that after I get finished taking voltage readings on my working board. I want to get that done first so I can have a good reference point to work from.
Blessings, Terry
Lol - It wasn't me that said max resistance first - blame Leach - he was the first one to start using that confusing phrase in the context of this amp. Then it was perpetuated by that master of smoke & mirrors Arius 😉 I was just following their lead 😀
Praise the Lord!
It's working.
I pulled the right channel out today so I could make a voltage chart and when I got it hooked up the rail fuses blew. I checked and had a couple of shorted outputs.
I just replaced all of them and the drivers just to be safe. Well I always like to use some 100R resistors when I first bring up the power on my variac so I can watch the voltage drop across them. Well, as I brought up the voltage I noticed a very high drop across the resistors. At full voltage I had 21V drop.😱
I tried backing off the bias pot and the voltage drop reduced. I should have figured that it was going to change when I swapped out the outputs but I hadn't. Anyway, I rest the bias and took measurements. They were pretty close to what Jens’ had posted except for the fact that my front end is running at about 68V from the regulator. I had 1.1V at the base of the drivers, which I figured is what really mattered.
Well, armed with this info, I backed off the bias pot on the channel I was having trouble with and was able to bring everything up to normal.
I reset the bias and all is well. I have a little more DC offset than before but it is still only 15mV so I think I'm fine.
I'm listening to it right now as I type and it sounds as lovely as I remember.
I can't tell you guys how much I appreciate all of you kind help. I hope some others can benefit from what I went through.
Blessings, Terry
It's working.
I pulled the right channel out today so I could make a voltage chart and when I got it hooked up the rail fuses blew. I checked and had a couple of shorted outputs.
I just replaced all of them and the drivers just to be safe. Well I always like to use some 100R resistors when I first bring up the power on my variac so I can watch the voltage drop across them. Well, as I brought up the voltage I noticed a very high drop across the resistors. At full voltage I had 21V drop.😱
I tried backing off the bias pot and the voltage drop reduced. I should have figured that it was going to change when I swapped out the outputs but I hadn't. Anyway, I rest the bias and took measurements. They were pretty close to what Jens’ had posted except for the fact that my front end is running at about 68V from the regulator. I had 1.1V at the base of the drivers, which I figured is what really mattered.
Well, armed with this info, I backed off the bias pot on the channel I was having trouble with and was able to bring everything up to normal.
I reset the bias and all is well. I have a little more DC offset than before but it is still only 15mV so I think I'm fine.
I'm listening to it right now as I type and it sounds as lovely as I remember.
I can't tell you guys how much I appreciate all of you kind help. I hope some others can benefit from what I went through.
Blessings, Terry
Glad to hear you got it running again.
Sorry if I confused you with setting the bias pot to max resistance for min bias. I don't have this particular board so it was safest to mention the resistance setting rather than the pot/preset direction.
The debug strategy that I had in mind was....
1) Going simple - no protection (not replace them, but remove them). Reset all trimmers (DC offset to mid value, if any, bias to minimum current).
2) Going basic - no input/output, measure voltages at output stage supply, input stage supply, input stage zener ref, VBE multiplier. Basically determine the DC quiescent states.
If this sounds like smoke and mirrors, I don't know how else to help.
PS: If you have no interest in Class A bias, you can crank up the value of R32 to prevent accidental gross overbias.
Have fun.
Sorry if I confused you with setting the bias pot to max resistance for min bias. I don't have this particular board so it was safest to mention the resistance setting rather than the pot/preset direction.
The debug strategy that I had in mind was....
1) Going simple - no protection (not replace them, but remove them). Reset all trimmers (DC offset to mid value, if any, bias to minimum current).
2) Going basic - no input/output, measure voltages at output stage supply, input stage supply, input stage zener ref, VBE multiplier. Basically determine the DC quiescent states.
If this sounds like smoke and mirrors, I don't know how else to help.
PS: If you have no interest in Class A bias, you can crank up the value of R32 to prevent accidental gross overbias.
Have fun.
- Status
- Not open for further replies.