Okay guys, my 300DC has been on the fritz for some time now and I was so embarrassed by what I did to break it, that i couldn't bear to seek help, but here goes.
I decided to test some power transistors (2SB645s and 2SD665s) that I suspected to be bad by swapping them into the left channel of my 300DC. My thinking was, if the transistors were bad, the protection relay would simply not click-on and I'd now if the transistors were bad.
The relay didn't click-on and nothing smoked, so I threw out the transistors and swapped the good ones in. The protection relay still stays turned off though. I left the amplifier on for a little bit to see if anything would smoke. Q711 awarded me with a nice POP and some smoke.
I don't have the service manual for the 300dc, so I don't know enough to identify the part to replace it, nor why it would blow.
I just need some help with how to proceed with getting the channel fixed. I'm quite a novice when it comes to fixing my own equipment beyond identifying blown components and poor solder joints. Thanks guys.
I decided to test some power transistors (2SB645s and 2SD665s) that I suspected to be bad by swapping them into the left channel of my 300DC. My thinking was, if the transistors were bad, the protection relay would simply not click-on and I'd now if the transistors were bad.
The relay didn't click-on and nothing smoked, so I threw out the transistors and swapped the good ones in. The protection relay still stays turned off though. I left the amplifier on for a little bit to see if anything would smoke. Q711 awarded me with a nice POP and some smoke.
An externally hosted image should be here but it was not working when we last tested it.
<HOTLINKED, SORRY>I don't have the service manual for the 300dc, so I don't know enough to identify the part to replace it, nor why it would blow.
I just need some help with how to proceed with getting the channel fixed. I'm quite a novice when it comes to fixing my own equipment beyond identifying blown components and poor solder joints. Thanks guys.
mechanicalman said:
Your first mistake. This is VERY poor troubleshooting practise.
Knowing there is a problem and waiting for the magic smoke to release?
The best advice and help on how to proceed I could give after reading all the above is to find yourself a good AUDIO TECH. IMO this amp is VERY MUCH worth having repaired by a pro and definitely not something I'd recommend you be learning on.
i'm very nervous about shipping it.
i know it's done, but the thing weighs close to 70lbs and the is cosmetically a 9/10.
i'm pretty sure any music store i take it to around here wouldn't take the time to match semi-conductors and would simply affect repairs i could do myself. i am very proficient with a soldering iron
btw, i've since rigged up my Hfe capable multimeter to a custom made socket, so i can test the transistors without ruining more 300DC amps.
i'm still confident that this problem isn't beyond my abilities if someone will help me identify what could be wrong with the left channel. i had a similar protection problem in my 1300dc that i traced to a faulty cap in the right channel (c723).
and i LOVE that 2130 so much. i never listen to the radio for anything anymore, but i promise you, it's the LAST component i'll ever part with.
i know it's done, but the thing weighs close to 70lbs and the is cosmetically a 9/10.
i'm pretty sure any music store i take it to around here wouldn't take the time to match semi-conductors and would simply affect repairs i could do myself. i am very proficient with a soldering iron
btw, i've since rigged up my Hfe capable multimeter to a custom made socket, so i can test the transistors without ruining more 300DC amps.
i'm still confident that this problem isn't beyond my abilities if someone will help me identify what could be wrong with the left channel. i had a similar protection problem in my 1300dc that i traced to a faulty cap in the right channel (c723).
and i LOVE that 2130 so much. i never listen to the radio for anything anymore, but i promise you, it's the LAST component i'll ever part with.
the q711 transistor is labled a977-r. i would assume that's an 2sa977-r.
after poking around here and looking at some datasheets, i was going to replace it with either a 2sa1209 and replace the complimentary PNP with a 2sa1211 (q712).
i was also considering shoe-horning a 2sa1535a (220 package) and a 2sc3944a pair. i'm having trouble locating somebody with the 2sc3944a though.
i'm pretty sure at this point, that the cause of the damage was a shorted driver (q003 or q004).
after poking around here and looking at some datasheets, i was going to replace it with either a 2sa1209 and replace the complimentary PNP with a 2sa1211 (q712).
i was also considering shoe-horning a 2sa1535a (220 package) and a 2sc3944a pair. i'm having trouble locating somebody with the 2sc3944a though.
i'm pretty sure at this point, that the cause of the damage was a shorted driver (q003 or q004).
The 2SA1209 looks like a good sub.
I assume, since you have an intact right channel, that you are comparing resistance measurements between to two channels to find other likely damaged components. There is a decent amount of circuitry between the bad outputs you were testing and Q711, and all of it's suspect at this point.
Cheers,
Paul
I assume, since you have an intact right channel, that you are comparing resistance measurements between to two channels to find other likely damaged components. There is a decent amount of circuitry between the bad outputs you were testing and Q711, and all of it's suspect at this point.
Cheers,
Paul
no, i hadn't actually considered investigating further. the rest of the board looked pristine and the back of the board didn't have any cooked pathways (it looks downright pretty).
i'll make sure to start comparing things before reinstall the board.
i'd kind of rather use the 2sa1535a/2sc3944a pair, just because the heatsink for q711/q714 tends to get really hot. i feel like a 220 package would do better than a 126 style transistor with all that heat. do you think there'd be anything wrong with using this pair as opposed to the drop-in replacement (2sa1209/2sa1211)?
i'll make sure to start comparing things before reinstall the board.
i'd kind of rather use the 2sa1535a/2sc3944a pair, just because the heatsink for q711/q714 tends to get really hot. i feel like a 220 package would do better than a 126 style transistor with all that heat. do you think there'd be anything wrong with using this pair as opposed to the drop-in replacement (2sa1209/2sa1211)?
against what big red will say .....
i have a couple of suggestions ....but of course this depends on your skills ......
There is an alternative way to do measuring when you have one channel working and one not .....usually when you measure things you use a multi meter for example to rail voltage and ground ...... to find out what your rail voltage is .....
well not in this case..... at this time you go to the collector of transistor bla of left working chanel to the same exact spot of the collector of transistor bla bla on the right not working chanel
the indication you get should be zero or with in a few millivolts .... big diference of 5-10-20 volts will point to you that the particular area has some kind of trouble .....
focuson protection circuits , seperate the protection from the amplifier and start trouble shooting the amp as is ( forget that your amp uses protection for a minute cause the fault might be on the protection circuits
a good start will be also to verify all the voltages mentioned in your schematic if something is wrong there this is also a fault indication
finally you may wana take a look here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=136261
happy regards sakis
i have a couple of suggestions ....but of course this depends on your skills ......
There is an alternative way to do measuring when you have one channel working and one not .....usually when you measure things you use a multi meter for example to rail voltage and ground ...... to find out what your rail voltage is .....
well not in this case..... at this time you go to the collector of transistor bla of left working chanel to the same exact spot of the collector of transistor bla bla on the right not working chanel
the indication you get should be zero or with in a few millivolts .... big diference of 5-10-20 volts will point to you that the particular area has some kind of trouble .....
focuson protection circuits , seperate the protection from the amplifier and start trouble shooting the amp as is ( forget that your amp uses protection for a minute cause the fault might be on the protection circuits
a good start will be also to verify all the voltages mentioned in your schematic if something is wrong there this is also a fault indication
finally you may wana take a look here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=136261
happy regards sakis
To me, Sakis's suggestion makes sense if/when you get to the point of the left channel's operation being non-self-destructive. The concern I have at this point is that the bad outputs have probably damaged other components that will cause Q711 to smoke again if it's the only thing replaced.
By doing power-off comparisons of resistances and diode check voltage drops you have a much better chance of finding and replacing the other damaged components before applying power.
By doing power-off comparisons of resistances and diode check voltage drops you have a much better chance of finding and replacing the other damaged components before applying power.
That's not the way I read it at all. What I read of the suggestion was to power on the amp and compare between like points in the two channels, e.g. one probe on base of 1st NPN output on left channel with other probe on base of 1st NPN output on right channel, and look for differences.
Anyway, sounds like you're proceeding reasonably. Hope it goes well.
Cheers,
Paul
Anyway, sounds like you're proceeding reasonably. Hope it goes well.
Cheers,
Paul
Hi mechanicalman,
Nice gear. My tuner is only a 2120, but I do have the 3650 and 300DC.
Now for things you should know. The transistors used also set the bias current. You need to be extremely careful with the protruding components on the rear, don't bent them back and forth at all. The traces may be delicate on the parts on the heat sink. Also, the voltage amplifier parts run very warm.
To redo solder joints, you need to remove the old solder and clean the leads mechanically. A little flux and new solder should do it. Don't use too much solder! Your speaker relay probably needs replacing. Digikey should have it, Mouser may as well.
Things about the design of this amp. Each amplifier has it's own high current power supply. Each voltage amp stage has it's own regulated power supply as well. That's four power supply sections and eight voltages, four regulated.
Now, what Bigred said about switching parts is very true. This is also an amplifier that has extremely good performance, but doesn't work like your average amplifier. The entire output stage depends on the transistor types as I have mentioned. It does not have a bias control section like other amps do. The thermally dependent resistors may be hard to find, I have never had to do that. Same for the diodes in similar positions.
Too bad I'm so far away, I've rebuilt a number of these. Each one with care and attention to detail. The results are worthwhile. If I knew anyone who was known to me personally in your area, I'd direct you there. Watch out for modifier type flakes. Your amp is a very good design. Part changes should be done with caution, and you don't need naked bulk film resistors or any other silly part changes. What you need is a careful technician who pays attention to what they are doing. An older Marantz technician might be good. Just avoid the young guys with stars in their eyes and too much enthusiasm.
-Chris
Nice gear. My tuner is only a 2120, but I do have the 3650 and 300DC.
Now for things you should know. The transistors used also set the bias current. You need to be extremely careful with the protruding components on the rear, don't bent them back and forth at all. The traces may be delicate on the parts on the heat sink. Also, the voltage amplifier parts run very warm.
To redo solder joints, you need to remove the old solder and clean the leads mechanically. A little flux and new solder should do it. Don't use too much solder! Your speaker relay probably needs replacing. Digikey should have it, Mouser may as well.
Things about the design of this amp. Each amplifier has it's own high current power supply. Each voltage amp stage has it's own regulated power supply as well. That's four power supply sections and eight voltages, four regulated.
Now, what Bigred said about switching parts is very true. This is also an amplifier that has extremely good performance, but doesn't work like your average amplifier. The entire output stage depends on the transistor types as I have mentioned. It does not have a bias control section like other amps do. The thermally dependent resistors may be hard to find, I have never had to do that. Same for the diodes in similar positions.
Too bad I'm so far away, I've rebuilt a number of these. Each one with care and attention to detail. The results are worthwhile. If I knew anyone who was known to me personally in your area, I'd direct you there. Watch out for modifier type flakes. Your amp is a very good design. Part changes should be done with caution, and you don't need naked bulk film resistors or any other silly part changes. What you need is a careful technician who pays attention to what they are doing. An older Marantz technician might be good. Just avoid the young guys with stars in their eyes and too much enthusiasm.
-Chris
Hi sakis,
This particular amplifier is more complicated than normal, and it's a classic piece. It's design and performance still rates highly. Care should be taken. When someone doesn't have the skills or equipment to handle a job like this, encouragement to forage ahead is not in anyone's best interest.
However, if that same person is constructing a new kit, go for it!
Within reasonable limits, yes.
This particular amplifier is more complicated than normal, and it's a classic piece. It's design and performance still rates highly. Care should be taken. When someone doesn't have the skills or equipment to handle a job like this, encouragement to forage ahead is not in anyone's best interest.
However, if that same person is constructing a new kit, go for it!
I have already prototyped it but would love to see/know more,
It is similar to the krill, but far simpler and nowhere as complicated
as the millet boosted bias diamond (tried that ,too.)
Soundwise, it is fast like the krill but does not seem to exhibit
the tendancy to oscillate at HF (5mhz+) . I did not use those
(q723/22) thermal resistors but opted for a standard Vbias
instead. Just for a baseline I wish someone would at least
post a BOM for this amp.(maybe a clearer schema as well)
thanks, OS
PS.. relieved that I didn't get "sinbinned" for voicing my opinion.
Hi ostripper,
That amplifier has been with me for an extended period of time. It has bested many other amplifiers over time. Since I had recently rebuilt it, and did some detail work, the sound quality has increased. One thing that completely disappeared was the slight "edginess". A few people who preferred a tube amplifier to this one noted that they no longer could find anything to complain about. I agree completely, cautiously since I'm the one who did the work. I really, really like this amplifier. Darn thing is pretty as well. Heavy as sin for it's physical size.
When you made your prototype, did you supply the voltage amplifier stage from it's own regulated power supply, a little higher than the main supply voltages? Also, the outputs should be the newer On Semi ring emitter types. They sound the closest to the original parts. If you used the MJ1502x parts, it will sound okay, but the MJ2119x types sound smoother, closer to the original design. Of course I have the advantage of having a 300DC to compare directly with. Did you also use a differential J Fet pair?
I have also built the output stage without using the thermistors. Like a classic diamond buffer, the transistors themselves provide bias control. I find that amplifier has a very slight thermal runaway. It disappears completely when a signal is run to a load. That tells me that your bias circuit has a slight amount of control.
If you look at a Nakamichi 620 power amplifier, you'll see that it is similar to the Marantz 300DC. Sounds similar too.
-Chris
That amplifier has been with me for an extended period of time. It has bested many other amplifiers over time. Since I had recently rebuilt it, and did some detail work, the sound quality has increased. One thing that completely disappeared was the slight "edginess". A few people who preferred a tube amplifier to this one noted that they no longer could find anything to complain about. I agree completely, cautiously since I'm the one who did the work. I really, really like this amplifier. Darn thing is pretty as well. Heavy as sin for it's physical size.
When you made your prototype, did you supply the voltage amplifier stage from it's own regulated power supply, a little higher than the main supply voltages? Also, the outputs should be the newer On Semi ring emitter types. They sound the closest to the original parts. If you used the MJ1502x parts, it will sound okay, but the MJ2119x types sound smoother, closer to the original design. Of course I have the advantage of having a 300DC to compare directly with. Did you also use a differential J Fet pair?
I have also built the output stage without using the thermistors. Like a classic diamond buffer, the transistors themselves provide bias control. I find that amplifier has a very slight thermal runaway. It disappears completely when a signal is run to a load. That tells me that your bias circuit has a slight amount of control.
If you look at a Nakamichi 620 power amplifier, you'll see that it is similar to the Marantz 300DC. Sounds similar too.
-Chris
If you can share anything, it would be greatly appreciated,
(schemas, drawings,anything).
Not directly, but I did use current sources (like the krill)
for the diamond, (see attached schema)
and a classic Vbias
with my standard OP stage (frugal based). It sounds great,
but I like complete info..
As far as a input stage I used my frugal3 holman (apt)CFP
differential to drive the diamond.
I have tinkered with this diamond thing forever ,after seeing
the krill and marantz, will not give up..
OS
Attachments
Hi ostripper,
That looks very close to what I built, minus the bias network.
My findings were that it sounds pretty good, promising. At higher power levels into a 4 ohm load, it sounds compressed but retains that wonderful detail.
This is what I was working on at the time of my accident. Now I don't completely understand my own work. It's like someone else actually did it, and I only have their notes to go from. That is both frustrating and a little scary.
The rest of the amp is pretty straight forward. A pair of very normal voltage regulators for the front end and a standard delay and DC offset protection network. You now have all the information you need. I may scan some of my notes at some time. This requires that I have both the energy and that I don't forget to do it.
Since this output stage is a single "drive line", it's pretty much completely self contained. You can drive it with whatever voltage amp stage that suites your fancy. My input impedance is pretty high, so I use a smallish coupling capacitor (0.22 uF?). DC offset hangs around 30 mV with no feedback, so I'm not really concerned about that right now. It has been driven by both tubes, op amps and various semiconductor voltage amplifiers.
It seems like a very nice building block. Interestingly, I went down this path after testing a signal diamond buffer that merrily drove 8 ohm speakers with very low distortion. The sound quality is very good, so I made it bigger.
That's how I arrived to this point.
-Chris
That looks very close to what I built, minus the bias network.
My findings were that it sounds pretty good, promising. At higher power levels into a 4 ohm load, it sounds compressed but retains that wonderful detail.
This is what I was working on at the time of my accident. Now I don't completely understand my own work. It's like someone else actually did it, and I only have their notes to go from. That is both frustrating and a little scary.
The rest of the amp is pretty straight forward. A pair of very normal voltage regulators for the front end and a standard delay and DC offset protection network. You now have all the information you need. I may scan some of my notes at some time. This requires that I have both the energy and that I don't forget to do it.
Since this output stage is a single "drive line", it's pretty much completely self contained. You can drive it with whatever voltage amp stage that suites your fancy. My input impedance is pretty high, so I use a smallish coupling capacitor (0.22 uF?). DC offset hangs around 30 mV with no feedback, so I'm not really concerned about that right now. It has been driven by both tubes, op amps and various semiconductor voltage amplifiers.
It seems like a very nice building block. Interestingly, I went down this path after testing a signal diamond buffer that merrily drove 8 ohm speakers with very low distortion. The sound quality is very good, so I made it bigger.
That's how I arrived to this point.-Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.