Correct, however Mooly, i know a couple of techs that would bias their amps only reading off the BE voltage drop... 0.54 and below , is too low, 0.55-62 was the optimal range they would set it at...
This is when they did not have a calculator handy to convert the mA's dropped across the ER when they measured the mV's... 😀
Are you watching the Windies/Saffer match right now?
Last edited:
Question. To make reference to the circuit easier can you tell me which channel is the faulty one. Its the one with no voltage across the 0.47 ohms so are they Rm73 and Rm 75 or Rm74 and Rm 76
I'm going to pick the first ones to describe what to measure next as I think you mentioned it was the left channel that was faulty. If its the other channel then you measure accordingly.
With the amp on measure the volt drop across Rm69 and Rm71. These are both 330 ohm and this should be a double check on your earlier reading of no base/emitter volts on that pair of outputs.
Now measure the base/emitter volt drop across both driver transistors Qm15 and Qm17. I can't find a respectable data sheet for these that reliably shows the pin outs but on the circuit diagram the base is the middle lead and the emitter is the one with the arrow... sure you knew that anyway 🙂 There should be around 0.7 volts across each.
It is a slightly odd fault you have tbh but measurement should reveal the cause.
You can also do the same base/emitter volt check across the reaming transistors in that channel, all should be around 0.7 volts with the PNP transistors (the ones where the arrow points in) having the emitter as the more positive with respect to the base.
The circuit has some key voltages marked, in particular the +1.2 and -1.2 volts on Qm 13 and the 5.9 volts where it say bias. You can do a quick check of all those voltages given.
And if non of that reveals anything, then we have to get really technical with it 😀
I'm going to pick the first ones to describe what to measure next as I think you mentioned it was the left channel that was faulty. If its the other channel then you measure accordingly.
With the amp on measure the volt drop across Rm69 and Rm71. These are both 330 ohm and this should be a double check on your earlier reading of no base/emitter volts on that pair of outputs.
Now measure the base/emitter volt drop across both driver transistors Qm15 and Qm17. I can't find a respectable data sheet for these that reliably shows the pin outs but on the circuit diagram the base is the middle lead and the emitter is the one with the arrow... sure you knew that anyway 🙂 There should be around 0.7 volts across each.
It is a slightly odd fault you have tbh but measurement should reveal the cause.
You can also do the same base/emitter volt check across the reaming transistors in that channel, all should be around 0.7 volts with the PNP transistors (the ones where the arrow points in) having the emitter as the more positive with respect to the base.
The circuit has some key voltages marked, in particular the +1.2 and -1.2 volts on Qm 13 and the 5.9 volts where it say bias. You can do a quick check of all those voltages given.
And if non of that reveals anything, then we have to get really technical with it 😀
ah, so sometimes the headphones are right after the driver stage, but it also can be done after the output stage?
And what stratus 46 meant was that the headphone was indeed after the output stage, but that only the driver stage was doing anything.
I didn't understood it correctly, I thought he meant that my headphones were directly after my driver stage. But he meant that only the driver stage was doing anything and the headphones are after the main output stage.
(the 4 transistors on the heatsink are my main output stage right?)
EDIT:
The ones without voltage across are rm73 and rm75 (left channel output)
Should I take a photograph?
Last edited:
Hi,
The correct bias is really determined by the topology (EF or CFP etc) and the component values used but anything from literally a milliamp upward will give "audibly perfect" sound and even zero bias in a correctly working amp will only give low level crossover distortion. It wouldn't sound like this one does if it were just lack of bias.
We have to gather all the evidence together first from taking readings 😀
(not a sports fan 🙂)
I will do it tomorrow, night shift again🙁.
What I find strange is that when I measured the resistance over the transistors on the heatsink, like you said.
Transistor 2 and 3 (both green ones) have no connection between center and b and center and e. And the black ones have resistance?
Is that because off PnP and NpN or anything like that?
What I find strange is that when I measured the resistance over the transistors on the heatsink, like you said.
Transistor 2 and 3 (both green ones) have no connection between center and b and center and e. And the black ones have resistance?
Is that because off PnP and NpN or anything like that?
Left channel it is then 🙂 I'll have to leave it for tonight but you have plenty to be going on with.
Yes, the four transistors are the outputs but the amp works as a whole, you can't really separate (and keep it working) the different stages. That said, the drivers (the transistors before the outputs) actually make a working amp even if you removed the output transistors but they would be very limited in the power they could supply because they feed the "speaker line" via those 330 ohms. That is what we suspect is happening although there seems a bit more going on here.
Measurements should hopefully reveal all, although we've plenty more options to attack this fault if needed.
Reading transistors in circuit doesn't always give good results on ohms ranges due to interaction with other parts. What we were looking for was shorts really where the meter would read really low like 0.03 etc.
NPN and PNP are different polarity devices. You should get a reading from base to collect and base to emitter. On diode range it will read around 0.700 that magic number again because the meter on the diode range actually reads the voltage across the test leads after injecting a small current through the device.
More tomorrow 🙂
General question
Hmm, but with the black transistors (output) I get a reading from C To B, and from C to E.
But with the green ones there's no connection between C to B and from C to E.
(with C is the middle pin)
with the blacks I indeed get 700 ohms!
I thought you told me to measure all transistors to center pin ( post 7)
But the most important, the new results.
no voltage over rm69 and rm71
qm15/ qm 17:
measuring from the middle pin to the one with the Arrow wasn't a succes (both left and right channel the only thing my meter shows is a 1.
But measuring the left and the right leg on the good channel says 0.565 Volt on both transistors.
But on the bad channel my meter goes from EDIT: was a fault, i am re measuring!
Hmm, but with the black transistors (output) I get a reading from C To B, and from C to E.
But with the green ones there's no connection between C to B and from C to E.
(with C is the middle pin)
with the blacks I indeed get 700 ohms!
I thought you told me to measure all transistors to center pin ( post 7)
But the most important, the new results.
no voltage over rm69 and rm71
qm15/ qm 17:
measuring from the middle pin to the one with the Arrow wasn't a succes (both left and right channel the only thing my meter shows is a 1.
But measuring the left and the right leg on the good channel says 0.565 Volt on both transistors.
But on the bad channel my meter goes from EDIT: was a fault, i am re measuring!
Last edited:
To make detailed resistive measurements on any of the transistors means removing them from the circuit.
However, with the red meter lead on the base, then for the NPN's you should read 0.700 between B and E and 0.7 between B and C.
For the PNP's they will read the same but this time keep the black lead on the base.
Any such readings need "interpretation" when measured in circuit. If there is any resisdual voltage around (from charged up caps) the reading can be influenced greatly and that could well be the cause of the 1 reading.
So have you no voltage across those 330 ohms (meter on volts now 🙂)
However, with the red meter lead on the base, then for the NPN's you should read 0.700 between B and E and 0.7 between B and C.
For the PNP's they will read the same but this time keep the black lead on the base.
Any such readings need "interpretation" when measured in circuit. If there is any resisdual voltage around (from charged up caps) the reading can be influenced greatly and that could well be the cause of the 1 reading.
So have you no voltage across those 330 ohms (meter on volts now 🙂)
Look at the first diagram showing the diode model. That is the basic test of the junctions and how they read.
Testing Transistors
Testing Transistors
But the most important, the new results.
no voltage over rm69 and rm71
qm15/ qm 17:
Measuring qm15 and 17 'wasn't that easy so I measured every thing...
The one with the Arrow I called 1:
qm15
1-3: 8.0v
2-3: 36v
1-2: 45V
qm17:
1-3: 10v
2-3: 36 V
1-2: 32V
Right channel
qm 18:
1-3: 0.58V
2-3: 36V
And well after that i mashed it up, i made a short circuit on the right channel while measuring qm18 1 and 2... Fuse is gone and I think so is my main output transistor 🙁..
no voltage over rm69 and rm71
qm15/ qm 17:
Measuring qm15 and 17 'wasn't that easy so I measured every thing...
The one with the Arrow I called 1:
qm15
1-3: 8.0v
2-3: 36v
1-2: 45V
qm17:
1-3: 10v
2-3: 36 V
1-2: 32V
Right channel
qm 18:
1-3: 0.58V
2-3: 36V
And well after that i mashed it up, i made a short circuit on the right channel while measuring qm18 1 and 2... Fuse is gone and I think so is my main output transistor 🙁..
Its all part of the learning process I'm afraid. You have to be careful measuring. Often its easier measuring to components that connect to the part you have trouble getting to.
Anyway, which fuse has blown ?
Where you go from here is really up to you. The parts to do a rebuild of both channels, thats outputs, drivers, probably the 0.47 ohms and few other bits are minimal in cost.
A lot depends on whether you feel confident working on it 🙂 And whatever the outcome its a good project to learn on.
Anyway, which fuse has blown ?
Where you go from here is really up to you. The parts to do a rebuild of both channels, thats outputs, drivers, probably the 0.47 ohms and few other bits are minimal in cost.
A lot depends on whether you feel confident working on it 🙂 And whatever the outcome its a good project to learn on.
Well when I did your test from post 7 (with the black lead Always at the centre pin or all of the 4 transistors). I Always measured from the pin in the center to the left and from the centre pin to the right.
But you say: measure from B to C and from B to E
But on the circuit bord there is B -C - E
So I measured from the C pin to B and from the C pin to E
Is there any change on maybe making the right channel not attached to the power suply, fuse keeps blowing and I would like to fix left first..
About the right channel:
The 0.47 ohms give the same value as before, but
q18 now has a short circuit between legs 2 and 3
And the green big transistor q22 has now a short circuit between
all the legs
and q20 has now a value of 95 ohms between the B and E leg (printed on the circuit bord)
Well it's a fun project and as you say, I just learn from it!
edit: the main fuse right next to the transformer: between pin 18 and 19
The 0.47 ohms give the same value as before, but
q18 now has a short circuit between legs 2 and 3
And the green big transistor q22 has now a short circuit between
all the legs
and q20 has now a value of 95 ohms between the B and E leg (printed on the circuit bord)
Well it's a fun project and as you say, I just learn from it!
edit: the main fuse right next to the transformer: between pin 18 and 19
Last edited:
Yes, in post #7, that was the start of faultfinding and suspecting a failed (shorted) output transistor at that stage seemed the most likely culprit. Testing from collector to emitter and collector to base was the first basic check... and it passed those at the time 🙂
Yep, we can build the left channel up first.
Best thing is to draw up a list of replacement semiconductors and other bits that are going to be needed. You'll need solder braid to remove the devices cleanly and a good iron with quite a large tip.
I'll look in again later 🙂
Yep, we can build the left channel up first.
Best thing is to draw up a list of replacement semiconductors and other bits that are going to be needed. You'll need solder braid to remove the devices cleanly and a good iron with quite a large tip.
I'll look in again later 🙂
Hopefully we'll get a good result from this. If your interested in seeing how a faultfinding process works then have a look at this,
http://www.diyaudio.com/forums/solid-state/233880-help-repairing-pioneer-m3.html
Your amps nothing like as complex as this one.
http://www.diyaudio.com/forums/solid-state/233880-help-repairing-pioneer-m3.html
Your amps nothing like as complex as this one.
ah, but in post #29, you are talking about resistance (ohm) measurement between B and E and B and C. (and not from C to B and from C to E)
And on the website you gave me they also say you need to measure from Base to Collector and from Base to emitter, and not from Collector to Base and Collector to emitter.
So that's why I am a bit confused, I don't get the difference of what we are doing between post 7 and 29
But ok, now we still don't know what's the problem with the Original bad channel, only that there are only high voltage over leg 1-3 on qm 15 and 17 where there should be 0.7 volt.
Edit:
thnx for the link, will be reading it soon, I think I really found a new hobby !
And on the website you gave me they also say you need to measure from Base to Collector and from Base to emitter, and not from Collector to Base and Collector to emitter.
So that's why I am a bit confused, I don't get the difference of what we are doing between post 7 and 29
But ok, now we still don't know what's the problem with the Original bad channel, only that there are only high voltage over leg 1-3 on qm 15 and 17 where there should be 0.7 volt.
Edit:
thnx for the link, will be reading it soon, I think I really found a new hobby !
Last edited:
Hmm can't let it rest, with the amp on you told me that the voltage drop is 0.7 volt. But testing on ohm range (diode range,) I get also 0.700 (7ohm00 ohm) is there a connection between these valeus or is that a coincidence that they are the same
It should be blowing fuses by now if the greenie is shorted out... the white ER next to it also looks fried.
Post #7 was a quick check of the most likely suspects... which proved at the time that they could well be OK (before the little mishap)
You can't hope to learn all this in a week, a month, a year... its a continuous learning process... for me it was the day job too 😀
700 millivolts (aprox) is the voltage at which a silicon junction begins to conduct. If you had a variable power supply that could be set from 0 through to say 10 volts and connected it across a diode or transistor junction (with the correct polarity to forward bias it) then as the voltage rise from 0 to say 0.5 volts, no current flows. Get to 0.6 and a small current flows. Get to 0.65 and perhaps 100 milliamps flows. Increasing this voltage now by even a fraction causes a massive increase in current and the junction will "clamp" the voltage never allowing more than around 0.7 to develop across it.
Look at figure 5 here,
Diode - Wikipedia, the free encyclopedia
So the "coincidence" with your meter showing around 0.7 and this magic 0.7 number is as I mentioned earlier. Your meter on the diode range has a voltage of probably around 3 volts across the open terminals. When you connect it across a diode the diode clamps that voltage to 0.7 and the meter is actually reading the millivolts between the meter terminals.
- Status
- Not open for further replies.