What do you mean by the little driver ? 
Qe17 and 19 need replacing. If the outputs are reading OK (not short) which they seem to be, then we'll go with just replacing the drivers and slowly bringing the supply voltage up again as a test. Refer back to where we first powered it up with the bias preset set at max and checking for no DC offset on the speaker output.
Qe17 and 19 need replacing. If the outputs are reading OK (not short) which they seem to be, then we'll go with just replacing the drivers and slowly bringing the supply voltage up again as a test. Refer back to where we first powered it up with the bias preset set at max and checking for no DC offset on the speaker output.
If it is easy to remove (or at least unsolder) the outputs then we can actually run the amp up without those output transistors fitted.
That might be an approach worth trying because it removes doubt over the outputs.
If you want to try that then let me know and I'll tell you what to measure when its powered up.
That might be an approach worth trying because it removes doubt over the outputs.
If you want to try that then let me know and I'll tell you what to measure when its powered up.
okay, will do. SWMBO is stirring. that marks the end of this mornings sojourn for me LOL.
i'll type you back same time tomorrow, with the new transistor drivers in place, and Qe23 reinstalled. why do you think Q17 smoked? it sounded so nice and clear. not quite so bassy as the Right channel, but clear as a bell.
aidan
i'll type you back same time tomorrow, with the new transistor drivers in place, and Qe23 reinstalled. why do you think Q17 smoked? it sounded so nice and clear. not quite so bassy as the Right channel, but clear as a bell.
aidan
no problem without them? would that stress Q17 and Q19?
sure, or i could install the new outputs. whichever you think is best. "easiest" doesn't matter, as i'm running out of time with this LOL
you're the boss, mooly. i've learned quite a bit the past few days. thank you so so much for your patience with me.
i'll check here a couple times throughout the day, and when i try to catch you tomorrow, i'll have the rig ready, as per your instructions.
best for now, and have a great day!
aidan
p.s. please say hello to sakis should you meet up with him today
OK
Qe17, well we needed to check a few things before turning it up. When the bulb is in place and the audio signal draws quite a bit of current, it causes the rail voltage to fluctuate and drop drastically. The bulb just can't supply that current and under those conditions the circuit can latch in an unpredictable state. That's why the volume must be kept really low. That's one possibility.
So two new drivers to be fitted. Bias set for minimum. And we could do with being certain those outputs are still good. If you remove or unsolder them then leave them that way and we'll power it up minus those being fitted. That will allow the bias generator and driver to be checked.
Qe17, well we needed to check a few things before turning it up. When the bulb is in place and the audio signal draws quite a bit of current, it causes the rail voltage to fluctuate and drop drastically. The bulb just can't supply that current and under those conditions the circuit can latch in an unpredictable state. That's why the volume must be kept really low. That's one possibility.
So two new drivers to be fitted. Bias set for minimum. And we could do with being certain those outputs are still good. If you remove or unsolder them then leave them that way and we'll power it up minus those being fitted. That will allow the bias generator and driver to be checked.
So here is the plan.
I've had another think on this and come up with some slightly different ideas. We will get this to work
1/ Isolate, and then test your output devices.
For Qe21 (the NPN) and Qe 23(the PNP) there should be no reading between C and E with the meter leads either way around. Do that test on a "high" ohms range (that is the range you might measure a 1 meg resistor on). That will show any leakage that the diode range may not. If you now switch the meter back to the diode check range you should read around 0.5 to 0.7 from B to E and from B to C on both devices. The NPN will read when the red lead is kept on the base, the PNP when the black lead is on kept on the base. Reverse the leads and there should be no reading.
If they both pass those basic tests we use them. The outcome of that depends on what we do next. Obviously if they fail then we have to fit the TIP outputs.
Question...... how many of the TIP devices did you get ? Was it one or two pairs ?
Also do you have any small low voltage electrolytic caps say around 10uf 16volt (or higher). I'll assume yes to that
3/ So depending on what you have parts wise, here's the plan...
If the original outputs are good... then I'm going to go with my instincts and suggest we get the TIP devices to function as the drivers replacing Qe17 and Qe19.
You also add the electrolytic across Qe7 between C and E. Just tag it on to the print under the device. Positive end to E (that's Re19) and negative end to the collector (Re23 end). The value of the cap is not critical, 1uf, 10uf, 47uf, all should work the same. Its function is to maintain a low impedance at AC between these two points and aid stability.
Bias preset set to 1k, and again we test by bringing the voltage up slowly. NO SPEAKERS attached. Assuming its OK you check the DC offset (voltage across the speaker terminals) and confirm its below 100mv. As before, use the bulb tester at all times.
If that is OK then its bias time again See if the bias adjusts... if it does not then again we check the voltage across Qe7 (same as across the new cap) and confirm it is actually altering with the preset. If it is then its the same routine as before. We drop the value of Re27 by degrees.
That's probably enough to be going on with for now. 10 minutes work tops
This shows the pin outs for the TIP41C (the TIP42C is identical) Be aware that the metal tab is internally connected to the collector so don't let it short to anything.
The other diagram shows the cap location.
The TIP41C replaces Qe17. The TIP42C replaces Qe19.
Check and double check that they are in correctly. Refer to my picture of the pinouts and confirm with your meter on ohms that each pin goes to the correct point on the circuit by measuring to one of the components it is supposed to go to.
I've had another think on this and come up with some slightly different ideas. We will get this to work
1/ Isolate, and then test your output devices.
For Qe21 (the NPN) and Qe 23(the PNP) there should be no reading between C and E with the meter leads either way around. Do that test on a "high" ohms range (that is the range you might measure a 1 meg resistor on). That will show any leakage that the diode range may not. If you now switch the meter back to the diode check range you should read around 0.5 to 0.7 from B to E and from B to C on both devices. The NPN will read when the red lead is kept on the base, the PNP when the black lead is on kept on the base. Reverse the leads and there should be no reading.
If they both pass those basic tests we use them. The outcome of that depends on what we do next. Obviously if they fail then we have to fit the TIP outputs.
Question...... how many of the TIP devices did you get ? Was it one or two pairs ?
Also do you have any small low voltage electrolytic caps say around 10uf 16volt (or higher). I'll assume yes to that
3/ So depending on what you have parts wise, here's the plan...
If the original outputs are good... then I'm going to go with my instincts and suggest we get the TIP devices to function as the drivers replacing Qe17 and Qe19.
You also add the electrolytic across Qe7 between C and E. Just tag it on to the print under the device. Positive end to E (that's Re19) and negative end to the collector (Re23 end). The value of the cap is not critical, 1uf, 10uf, 47uf, all should work the same. Its function is to maintain a low impedance at AC between these two points and aid stability.
Bias preset set to 1k, and again we test by bringing the voltage up slowly. NO SPEAKERS attached. Assuming its OK you check the DC offset (voltage across the speaker terminals) and confirm its below 100mv. As before, use the bulb tester at all times.
If that is OK then its bias time again
See if the bias adjusts... if it does not then again we check the voltage across Qe7 (same as across the new cap) and confirm it is actually altering with the preset. If it is then its the same routine as before. We drop the value of Re27 by degrees.That's probably enough to be going on with for now. 10 minutes work tops
This shows the pin outs for the TIP41C (the TIP42C is identical) Be aware that the metal tab is internally connected to the collector so don't let it short to anything.
The other diagram shows the cap location.
The TIP41C replaces Qe17. The TIP42C replaces Qe19.
Check and double check that they are in correctly. Refer to my picture of the pinouts and confirm with your meter on ohms that each pin goes to the correct point on the circuit by measuring to one of the components it is supposed to go to.
Attachments
morning all. mooly, sorry, but i've just installed new 5551 and 5401, and reinstalled the Qe23. both Qe21 and Qe23 tested good when pulled.
if you want, i'll pull the outputs (though its a bit of work to get them out and in again).
yes, plenty of caps, though probably much higher voltage rating.
i thought the TIPs were for output, no?
i bought 2 X TIP41 and 2 X TIP42, and 2 X 5551 and 5401 though i'm down to the last of those, as you so aptly referred to them as the "little'uns".
let me know which configuration you'd like.
i've already got the bias set at 1K resistance on both sides. speakers are connected.
tuner is installed at TUNER input.
aidan
if you want, i'll pull the outputs (though its a bit of work to get them out and in again).
yes, plenty of caps, though probably much higher voltage rating.
i thought the TIPs were for output, no?
i bought 2 X TIP41 and 2 X TIP42, and 2 X 5551 and 5401 though i'm down to the last of those, as you so aptly referred to them as the "little'uns".
let me know which configuration you'd like.
i've already got the bias set at 1K resistance on both sides. speakers are connected.
tuner is installed at TUNER input.
aidan
OK, as you have it all configured lets continue with what you have. Be absolutely certain that the 2N's are correctly orientated.
The speaker switch sounds worrying. Can you see it physically damaged ? As long as there is no short between output and ground or output to the other channels output then its OK (well OK as such that it wont cause harm).
Yes, the TIP's are output devices but can also be used as drivers. I did some more work on the simulation of the amp to see how it all looked............. we can come to that later if need be.
I would still fit the cap I mentioned even with the 2N drivers fitted.
Then its test as before. No speakers and just see if the offset is correct and if the bias adjusts OK.
The speaker switch sounds worrying. Can you see it physically damaged ? As long as there is no short between output and ground or output to the other channels output then its OK (well OK as such that it wont cause harm).
Yes, the TIP's are output devices but can also be used as drivers. I did some more work on the simulation of the amp to see how it all looked............. we can come to that later if need be.
I would still fit the cap I mentioned even with the 2N drivers fitted.
Then its test as before. No speakers and just see if the offset is correct and if the bias adjusts OK.
heres a tip for those who don't already know.
in the base of those new-fangled CFL light bulbs, you will find a 10 or 20uF 200V electrolytic cap, a couple of .1's and a couple of .01's. each manufacturer uses different component values, but all i have seen have the above. nice FREE source for caps.
thought people might be interested. AND, you don't have to break the glass. the base is press-fit in to place. just pry it loose, cut the two bulb leads, and you get the prize. kinda like a kinder-toy for geeks.
aidan
in the base of those new-fangled CFL light bulbs, you will find a 10 or 20uF 200V electrolytic cap, a couple of .1's and a couple of .01's. each manufacturer uses different component values, but all i have seen have the above. nice FREE source for caps.
thought people might be interested. AND, you don't have to break the glass. the base is press-fit in to place. just pry it loose, cut the two bulb leads, and you get the prize. kinda like a kinder-toy for geeks.
aidan
wow. not good.
in speakers "OFF", shows open from both L and R + terminals. in all position, all 4 grounds of speakers A and B show short to chassis
in speaker A position, only 7 ohms from both positive terminals of A to chassis gnd, and B channel shows open.
in B position, both of A channels positive terminals show 11 Kohm, B speakers shows 200 Kohm.
in A+B, both A and B speaker positive terminals show 11K to chassis. so, i suspect the switch. it looks like it melted a bit on the top.
aidan
in speakers "OFF", shows open from both L and R + terminals. in all position, all 4 grounds of speakers A and B show short to chassis
in speaker A position, only 7 ohms from both positive terminals of A to chassis gnd, and B channel shows open.
in B position, both of A channels positive terminals show 11 Kohm, B speakers shows 200 Kohm.
in A+B, both A and B speaker positive terminals show 11K to chassis. so, i suspect the switch. it looks like it melted a bit on the top.
aidan
Last edited:
no problem. i'll be puttering here a bit.
question for when you return. when the switch, in a functioning amp, is set to A speakers, is this the same as leaving B speaker terminals hanging? when in B position, are the A leads from the amp simply disconnected from the speakers, or is there a load introduced to mimic a speaker?
are the speaker leads from the amp, before it meets the switch on its way out, hot? i mean, are they always on?
take your time sir. i'll be waiting.
and no, i have not seen "spice light", i think you called it. free? do you have a link?
a
never mind. sometimes i'm such a "duhh"!
http://lmgtfy.com/?q=free+version+of+LTspice for those "special" guys like myself.
thanks mooly, i'm going to get it this morning. may (almost certainly will) have a question or 37 for you
aidan
http://lmgtfy.com/?q=free+version+of+LTspice for those "special" guys like myself.
thanks mooly, i'm going to get it this morning. may (almost certainly will) have a question or 37 for you
aidan
one way for people that don't have instruments to work with is to place a resistor of some wattage in series with the speaker something like 500 R or so ...
The idea is to be able to still listen music from the output but through the resistors the current draw from the load will be very very low so if there is mistakes you will probably find out before you blow anything
Its an unorthodox approach but some time helps ...Remember that initially all tests are without load ... if you have load connected and an amplifier that oscillates for any reason it will blow far before you put music to play
Kind regards
Sakis
The idea is to be able to still listen music from the output but through the resistors the current draw from the load will be very very low so if there is mistakes you will probably find out before you blow anything
Its an unorthodox approach but some time helps ...Remember that initially all tests are without load ... if you have load connected and an amplifier that oscillates for any reason it will blow far before you put music to play
Kind regards
Sakis
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.