I accidently slipped with the probe briefly connecting the negative rail to the base on Q621, collector on Q615 and emitter on Q609. The positive rail fuse blew as this happened. Changed the fuse but it blew as I powered up the amp.
Any insights as to what could have been damaged are very welcome.
Schematic below and the blue line is where the short happened and marking the positive rail fuse (T4).
Any insights as to what could have been damaged are very welcome.
Schematic below and the blue line is where the short happened and marking the positive rail fuse (T4).
Hi!
I'd check all the 6 output transistors and the 2 drivers. Chances are one of them blew due to the high current during the short.
Next time you power the amp up, use some resistor limiter in series with the mains (60W bulb or an equivalent resistor) - this helps to protect other circuit parts.
I'd check all the 6 output transistors and the 2 drivers. Chances are one of them blew due to the high current during the short.
Next time you power the amp up, use some resistor limiter in series with the mains (60W bulb or an equivalent resistor) - this helps to protect other circuit parts.
You need to understand the current paths available when the short occurred, locate obviously failed devices and then make a determination based on that as to what else may or may not have been impaired.
This is the fault current path via all the base/emitter junctions. Q621 and Q625 which are low powered devices have taken the full fault current. Also remember that current has come via the upper NPN's which may also have suffered.
Ultimately you will probably only find one or two failed devices but it then becomes a question of whether others have been impaired in any way. Only when you know the actual failures can you decide whether to go further and replace others.
This is the fault current path via all the base/emitter junctions. Q621 and Q625 which are low powered devices have taken the full fault current. Also remember that current has come via the upper NPN's which may also have suffered.
Ultimately you will probably only find one or two failed devices but it then becomes a question of whether others have been impaired in any way. Only when you know the actual failures can you decide whether to go further and replace others.
Ok, there are no obvious blown/burnt transistors so I would need to take them out and check them individually.
Are there any conclusions to draw from the fact that the positive rail fuse blows and not the negative, no speaker connected? I would think both fuses should break or is it just a timing issue with one breaking before the other.
Are there any conclusions to draw from the fact that the positive rail fuse blows and not the negative, no speaker connected? I would think both fuses should break or is it just a timing issue with one breaking before the other.
when you say "no obvious" do you mean by sight, or by using a DMM on diode setting to check the B-C-E of each of the transistors Mooly highlighted in the path? I would check with DMM as a first option, before pulling out all those transistors. Then you can remove 1 or 2 blown components and then you're able to confirm the short in the circuit path is gone with the DMM. Try to work methodically through the path Mooly provided and you should find the issue.
IMO the positive fuse blows because of the PNP short, full amplifier current is being directed to the NPN/positive rail and once it exceeds that rails fuse rating, it blows. Likely means you have full rail voltage on the speaker output, but you can't see/test it because you need to limit the current to keep the fuses intact, and not damage other components long enough to do the test.
IMO the positive fuse blows because of the PNP short, full amplifier current is being directed to the NPN/positive rail and once it exceeds that rails fuse rating, it blows. Likely means you have full rail voltage on the speaker output, but you can't see/test it because you need to limit the current to keep the fuses intact, and not damage other components long enough to do the test.
Don't forget the positive side was battling that current too. So include the positive rail parts including Q619.
Always replace one stage back from the last damaged part. One output means all outputs and drivers minimum. Check R637 and 639 along with the emitters.
Always replace one stage back from the last damaged part. One output means all outputs and drivers minimum. Check R637 and 639 along with the emitters.
Other ways and means that save desoldering lots of parts. First thing though is to use the DBT (bulb tester).
With the bulb in place switch on and the bulb should be bright with a fault. Measure the DC voltage across all six 0.22 ohm emitter resistors and see if any of the six have voltage across them. Record the voltage across each and report back with the result.
Get to that stage and lets see where its at.
At the end of the day its 95% certain its going to be failed drivers and perhaps one or more failed output transistors.
Hi,
When power transistors blow, as a result of a high current short, normally there is no external sign, specially when there are fuses. It's just a short inside of it.
With a multimeter, you'll be able to see if they are shorted even in the circuit.
Using multimeter in the "diode mode", check each junction Base-Emitter and Base-Collector.
In the forward direction (check if NPN or PNP), you should see 0.5V to 0.7V and in the reverse, open circuit, unless there are other components influencing - any doubt, post here.
But since most of times power transistors fail as a short, you are going to see 0ohms (or 0V in diode mode) forward and reverse.
Since there are transistors in paralell, once you find a short, then you need to isolate to see which one is shorted - You need to start isolating to check which one has failed.
The idea of using a current limiter (bulb or a resistor) and measure the current on the 0.22ohm resistors can also help to identify which one from the 3 paralell transistors is shorted.
Q619 isn't connected to the output, though. It would have to deliver the base current for Q623, but that's about it. Not thereby said that this current couldn't have been excessive before the fuse blew.
The top half of the output is likely beefy enough to allow the fuse to blow before the devices do. Then again, it would not be the first time a semiconductor device had sacrificed itself in an attempt to protect a rail fuse.
If I was doing the repair, and one half of the output stage was fried, I'd consider replacing all the output devices along with the drivers. The additional devices will cost less than the cost of shipping the devices to you.
Tom
Hi Tom,
That is exactly how I would approach it. Test to see if we lost an output or driver, go from there. Replace the entire stage if you have a shorted output or driver. Fuses are time delay opening devices. Maybe a short time for humans, but an eternity in semiconductor land.
Devices can go defective without actually shorting. Characteristics change and later they may well fail without warning.
That is exactly how I would approach it. Test to see if we lost an output or driver, go from there. Replace the entire stage if you have a shorted output or driver. Fuses are time delay opening devices. Maybe a short time for humans, but an eternity in semiconductor land.
Devices can go defective without actually shorting. Characteristics change and later they may well fail without warning.
Interesting is that the minus rail fuse refuses to pop. I would stick a lower rated one there out of curiosity if it survive again. If yes, there's obviously not rail to rail short, meaning there's some hope .. I know it's not likely, because where it would find the path to the ground, but hey, maybe you overlooked something trivial.
A fuse is a temperature thing. If one has a longer time constant, it would probably survive if the other went first. I would examine the element to see if it has a heated mark in the middle, and replace itf it showed any signs. Otherwise it may simply let go sometime later - or sooner. You just don't know.
Great help everyone, thanks! I understand the bulb thing but don't have one, not sure old style bulbs exist anymore but I'll see what I can find. I also understand I can use a resistor but don't have one with high enough value and power rating at hand.
I did the diod test in situ and as the other channel is good I could compare thinking I would be able to spot an open Q junction by getting a different reading on the faulty channel.
I found the positive rail output stage with Q627, 631 and 635 as well as Q625 and Q619 to be shorted (in purple below). R637, 639 are ok and Q613, 615 diod measure the same as the good channel.
Considering the advice to replace all in the stage if one has broken I would need to replace all the output/power transistors, both drivers Q625/623 and Q619/621?
Should I also replace Q613, 615 as they might have been stressed and possibly fail later?
I did the diod test in situ and as the other channel is good I could compare thinking I would be able to spot an open Q junction by getting a different reading on the faulty channel.
I found the positive rail output stage with Q627, 631 and 635 as well as Q625 and Q619 to be shorted (in purple below). R637, 639 are ok and Q613, 615 diod measure the same as the good channel.
Considering the advice to replace all in the stage if one has broken I would need to replace all the output/power transistors, both drivers Q625/623 and Q619/621?
Should I also replace Q613, 615 as they might have been stressed and possibly fail later?
Lightbulb in series with power supply serves as current limiter. Usually it's used on the primary of the transformer, but you as well can put it anywhere. Just voltage of the bulb should fit the supplying voltage and its wattage shouldn't be that high that it might introduce risky current. Principle is simple. When the bulb's filament is cold, it has low resistance. But when the short occurs in the circuit it causes the bulb to shine, then the resistance is high, current is according to the bulb wattage and the circuit is protected. You indeed need classic bulb.
As for the transistors replace, experienced craftsmen there will probably tell you to replace all ten and I'm recommending doing so too. Ideally also with Q613/615, and Q617 too for a peace of mind.
As for the transistors replace, experienced craftsmen there will probably tell you to replace all ten and I'm recommending doing so too. Ideally also with Q613/615, and Q617 too for a peace of mind.
Hi A 8,
The actual thing you need here is called a variac. It enables you to slowly increase the AC mains voltage in a controlled manner. That way you can avoid things going wrong and actually troubleshoot intelligently.
You might be able to find one used, or borrow one. These are valuable to have around if you do much with circuitry. For example you can use them to get a lower voltage out of a transformer so you can run a circuit before investing in the exact, correct transformer.
The problem with using a light bulb is that it changes it's resistance with current. That means as the current changes in your circuit, the voltage you are feeding it also is changing. This makes it a lot more difficult to do meaningful troubleshooting. The one and only reason these are recommended and popular is that they are cheap. Period. Circuit damage can still occur using the dim bulb type circuit.
The actual thing you need here is called a variac. It enables you to slowly increase the AC mains voltage in a controlled manner. That way you can avoid things going wrong and actually troubleshoot intelligently.
You might be able to find one used, or borrow one. These are valuable to have around if you do much with circuitry. For example you can use them to get a lower voltage out of a transformer so you can run a circuit before investing in the exact, correct transformer.
The problem with using a light bulb is that it changes it's resistance with current. That means as the current changes in your circuit, the voltage you are feeding it also is changing. This makes it a lot more difficult to do meaningful troubleshooting. The one and only reason these are recommended and popular is that they are cheap. Period. Circuit damage can still occur using the dim bulb type circuit.
Remember that the 6 output transistors are actually 3 parallel pairs and if one reads short it automatically means all read short in that upper or lower group. Isolating the 0.22 ohm emitter resistors will let you locate individual failures..
For test purposes the amp will fully function with any one good NPN and any one good PNP output in place.
The drivers and pre drivers should be replaced.
Q613 and 615 are most likely OK.
Be very careful where you get replacements from, and do not buy from eBay as a general rule.
For test purposes the amp will fully function with any one good NPN and any one good PNP output in place.
The drivers and pre drivers should be replaced.
Q613 and 615 are most likely OK.
Be very careful where you get replacements from, and do not buy from eBay as a general rule.
Yes, I understand but was under the impression that it was best practise to replace all if one had broken? So figured I'll simply get replacements and swap them all and not dive deeper into it. They are hard to find though so if I could get away with maybe changing one that would be easier (I have 2 pairs already).
Indeed, that seems to be the bigger issue right now, I found these guys Littlediode that have all that I need for this repair but there are a few older posts on here with some components being fakes. Anyone have any insight or experience of them during the last 2-3 years?