I'm about to order up the transistors for normal output but I'm not sure what to order since I have a mishmash installed in my D940. I'll just say "left" channel has a1386/c3519 and specs are (160v 15a 130w), "right" has a1386A/c3519a rated for (180v 15a 130w) but this part is unavailable and its replacement(a1294/c3263) is rated at (230v 15a 130w). Which would be the best set to go with? Hmm on closer inspection it appears somebody has already replaced at least one of the transistors..C3519, did a lousy job too, I'm assuming now that it came equiped with a C3519A's but since they are more expenive and harder to find the guy just chucked in the lower spec C3519. Think the a1294/c3263 would be a viable replacement?
S.Spielbergo said:
I think that's not going to be a problem, however, do try to match the transistors (NPN to PNP), and always measure before testing it on your speakers... (Or else I guess we might be talking about speaker coils next time 😀)
While waiting for the replacements, I would recommend just hooking up some wires on the output of the power stage before the relay, and see if there's anything funny there.
Hi S.Spielbergo,
Hi Maniac,
It's only audio, nothing is that big of a rush and there is no need to take risks that are not necessary.
Sorry if it seems I am picking on you. I'm only trying to correct some very poor practices and make sure no one else begins doing these things.
-Chris
"Struck by technician". Very common cause of further troubles. Those other replacement transistors should work.
Hi Maniac,
Don't do that. I've seen equipment come into my shop with teh relay still shorted out. Besides, there is no reason to do this. Just simply measure on the other side of the relay. No biggie.
I am. Most designs will have DC offsets below 50 mV and that's my limit. At 0.5 volts your DC offset protection may operate if it exists. I would say that the amplifier is defective or the design is defective.
No! Never do that! Never move transistors around to test. Any number of things may happen and the "DPO", or commutator transistors are sometimes different than the outputs. That is a very bad practice.
It's only audio, nothing is that big of a rush and there is no need to take risks that are not necessary.
Sorry if it seems I am picking on you. I'm only trying to correct some very poor practices and make sure no one else begins doing these things.
-Chris
For soldering a wire to test, I only recommend that because if you have that and some test lead with alligator clips. it will make testing rather easy and allows you to monitor the condition vs time with ease.
For DC components, I would regard 50mV to be defined as "Headphone grade" DC component guideline. I do however try to tune those stuff to below 5mV when I could, but even most headphones would be able to tolerate that much and more.
For speakers, most speaker protection device I have seen seems to have protection set at 2V or so of DC output.
For transistor "transplants", I only suggest that as a last ditch choice. DPD circuits are only used as switches to boost the max voltage swing. When no suitable replacement can be found, and I clearly stated below that statement you quoted that, this operation should only be done when you have tested the transistor itself for HFE level and matching status to be very close to the ones in the good channel, since he got two defective channels, this wouldn't work for him I suppose.
DPD transistors themselves can be easily replaced by pin compatible transistors that offer similar rating but does not require matching at all. Matching is a bear, and if the ones on DPD is already fairly close, it would actually be better than buying new transistors, as it is highly unlikely you'd get matched sets like you can with tubes.
For DC components, I would regard 50mV to be defined as "Headphone grade" DC component guideline. I do however try to tune those stuff to below 5mV when I could, but even most headphones would be able to tolerate that much and more.
For speakers, most speaker protection device I have seen seems to have protection set at 2V or so of DC output.
For transistor "transplants", I only suggest that as a last ditch choice. DPD circuits are only used as switches to boost the max voltage swing. When no suitable replacement can be found, and I clearly stated below that statement you quoted that, this operation should only be done when you have tested the transistor itself for HFE level and matching status to be very close to the ones in the good channel, since he got two defective channels, this wouldn't work for him I suppose.
DPD transistors themselves can be easily replaced by pin compatible transistors that offer similar rating but does not require matching at all. Matching is a bear, and if the ones on DPD is already fairly close, it would actually be better than buying new transistors, as it is highly unlikely you'd get matched sets like you can with tubes.
Well guys, even with both sets of of output transistors (L&R)desoldered and removed from the circuit the protection relay pops open shortly after powering up. I tried to get some voltage values from underneath the relay but with my digital VOM it was a little tough, 25v controling the relay drops to ~4v when it is open, at the other end where the speaker leads come in from the output devices and exit to the speaker terminals there was around .4v with the relay closed. 
I'm running out of ideas.
I'm running out of ideas.
Hi S.Spielbergo,
I think your 0.4 VDC is a clue, and far too high. Some DC offset protection circuits will trip lower than that.
Hi Maniac,
-Chris
I think your 0.4 VDC is a clue, and far too high. Some DC offset protection circuits will trip lower than that.
Hi Maniac,
I don't want to get into a war here, but your figures are at odds with what I see in the field and what has been recommended by many service managers.
So have I, but let's not accept the lowest common denominator here.
I have no trouble matching new transistors. Tubes ae more difficult, but I match those as well. The trick is to buy a bunch from teh same lot, then tubes vary more than transistors due. Especially some of the newer On Semi devices.
-Chris
anatech said:
Hi Chris,
I can only say that it would be interesting matching those Sanken stuff, especially with the lot of it on the market being old and possibly mixed stock.
Hi Maniac,
Use from the same lot numbers. If they are new, Japanese transistors have normally been good as far as parameter spread is concerned. I'd be using current parts anyway.
-Chris
Use from the same lot numbers. If they are new, Japanese transistors have normally been good as far as parameter spread is concerned. I'd be using current parts anyway.
-Chris
(swallows pride), I'll just ask...how do I go about matching tansistors? I'm still dumbfounded as to why there is too much juice passing through the relay and tripping it since the output devices are gone.
Hi S.Spielbergo,
I have posted about this in the SymAsym threads (one anyway) and a few others. I'll see if I can find them. Have a look on your own. I also posted a circuit that gives much better matches for testing.
If an output shorts, that can reverse bias one input transistor. That will degrade the performance and change the transistors characteristics. Your drivers may be connected to the output also, and you would have to pull those too. I would automatically replace those and the bias transistor if I found so much as one output gone.
-Chris
No need for that. Many people don't understand the concept.
I have posted about this in the SymAsym threads (one anyway) and a few others. I'll see if I can find them. Have a look on your own. I also posted a circuit that gives much better matches for testing.
If an output shorts, that can reverse bias one input transistor. That will degrade the performance and change the transistors characteristics. Your drivers may be connected to the output also, and you would have to pull those too. I would automatically replace those and the bias transistor if I found so much as one output gone.
-Chris
You've been getting some good advice here, but I'll jump in with my 2 cents worth.
I second pulling the drivers - it was probably repaired once before, it broke again, so do it right. I'd also take a good look at anything that is on the PCB were it looks like the fiberglass has darkened.
You can easily open up the holes from anything you pull off of the PCB with a wooden toothpick.
I second pulling the drivers - it was probably repaired once before, it broke again, so do it right. I'd also take a good look at anything that is on the PCB were it looks like the fiberglass has darkened.
You can easily open up the holes from anything you pull off of the PCB with a wooden toothpick.
Hanginon said:
Proton's single sided board is easy to open up, just pull the part out with the method you are comfortable with, and the mounting hole will open up. If not, an additional swipe from your soldering iron will make it so.
I can't seem to make any progress on this receiver at all. Every time I test a section of it I get nothing. I have reinstalled the original output transistors to their respective chanel, left I think. Tested the solder joints for continuity with an ohm meter and I still get nothing for output. I have a crappy speaker that was headed for the trash and have it connected via the speaker output upstream of the protection relay and selector switch and I get zippo. I guess if I can cut power to the amplifier section I can still use it as a preamp😡
Hi S.Spielbergo,
All right then. Let's start from the beginning then.
Are your outputs still hot? What DC voltage do you have on the outputs before the relay? Can you post a picture of this amp?
-Chris
Edit: The AM300 looks like a conventional amp, not the odd disaster that most NAD's were inflicted with. Should be a good unit once repaired.
All right then. Let's start from the beginning then.
Are your outputs still hot? What DC voltage do you have on the outputs before the relay? Can you post a picture of this amp?
-Chris
Edit: The AM300 looks like a conventional amp, not the odd disaster that most NAD's were inflicted with. Should be a good unit once repaired.
Ok, for starters the speaker terminals on the back appear to be inoperative, they provide neither a signal nor ground so I have hooked up to the output over by the relay and grounded to chassis, now we have sound! The channel with the proper output transistors is putting out ~4.8mv, give the transistor a little rap with the fingertip and voltage swings enough to trigger the relay open. Now the other channel with the jury rigged transistors is producing ~45mv consistently and sounds like rubbish. Both channels produce a residual output even when the unit has been switched off.
Hi S.Spielbergo,
Okay, we now know you have two possible problems. How close does the schematic posted earlier look to your amp?
-Chris
Okay, we now know you have two possible problems. How close does the schematic posted earlier look to your amp?
The relay cut out could be caused by a fault on either channel. You need to figure out which one. Use a voltmeter and try to figure out if you are getting a DC offset. You may have to try several times to get a reading. An intermittent junction in a power transistor will not cause this in most designs, the fault will be further back into the voltage amp section.
Your DC bias current is probably non-existent. Have you checked the emitter resistors and output transistors fully? Also test the driver transistors, I normally make sure the diff pair is running by checking the common emitter voltage. Check any supplies for the voltage amp stage.
-Chris
anatech said:
Actually, AM-300 I got does not have any problem I can find, I just want to mod the heck outta it. 😛 But it would be a PITA if I had to check every component value and trace the circuit myself. So I did what everyone else would do in this situation, get a service manual and just read the BOM off it, instead of trying to figure it out from the board 😛
Hi Maniac,
I can't imagine tracing the entire thing out. I do rough diagrams myself because there are only so many ways to build an amp. The weird ones need a manual.
-Chris
Absolutely!
I can't imagine tracing the entire thing out. I do rough diagrams myself because there are only so many ways to build an amp. The weird ones need a manual.
-Chris
I took a few more voltage readings this morning, for starters my dc readings at the speaker outputs have changed, the good side has gone way up to ~7mv (slowly falling from 37mv cold) from ~.4mv last night. The bad side continues to soldier on at a steady and appauling 38mv.I figured I'd check the output transistors one more time for inconsistancies and I got C3519A(.55v/38v/.9v), A1386A(.8v/38v/.38v) on one channel and C3519(.37v/38v/8.5v), A1386(9.2v/38v/9.5v) on the other channel. Now, the driver transistors(4 little guys with heat sinks) showed 54v/26v/54v, 54v/1v/54v on the first channel and 54v/29v/54v, 54v/10v/54v on the other channel. Anatech or Maniac, would either of you be able to just quickly desribe a flow chart of operations from preamp out to speaker outputs for me? Every ,time I try to find info on amplifier design its either grossly oversimplified or waay to advanced, mostly the former
Thanks guys
Thanks guys
Hi S.Spielbergo,
Can you please refer to the schematic posted earlier? This would really help.
-Chris
Can you please refer to the schematic posted earlier? This would really help.
Indicating pretty low bias current I think. You need to put your negative lead to the speaker output and reference that for each channel.
This shows a rather dramatic failure, I think. The four transistors in he center of the heat sink are there to switch to the higher voltage supply. Readings with respect to ground on each collector and emitter for those might be helpful. The two transistors straddling the small ones on the heat sink are your outputs. The small transistor is for bias control. The terminal voltages on these with respect to the speaker line would be helpful.
-Chris
- Status
- Not open for further replies.