Greetings,
I just picked up a McIntosh MC2100 for an embarrassingly small amount of money. Obviously it is not in working order, it blows the 5amp fuse immediately upon plugging it in. My forte is vacuum tube amps so fixing this will be a learning experience in many respects. I'm going to recap the whole thing and believe that the cause of the blown fuses are the output transistors. My reading has led me to believe that all 12 power transistors can be replaced with one type of transistor, MJ15003G. My question is this. In my amp I have 4 different part numbers on my output transistors, does the MJ15003G replace all 12 output transistors? And any tips, tricks or words of encouragement would be greatly appreciated.
I just picked up a McIntosh MC2100 for an embarrassingly small amount of money. Obviously it is not in working order, it blows the 5amp fuse immediately upon plugging it in. My forte is vacuum tube amps so fixing this will be a learning experience in many respects. I'm going to recap the whole thing and believe that the cause of the blown fuses are the output transistors. My reading has led me to believe that all 12 power transistors can be replaced with one type of transistor, MJ15003G. My question is this. In my amp I have 4 different part numbers on my output transistors, does the MJ15003G replace all 12 output transistors? And any tips, tricks or words of encouragement would be greatly appreciated.
I don't have an MC2100 schematic but I've fixed an old amp or two. MC15003g is a 1990? technology TO3 transistor which is still for sale but costs extra. If you're going to replace all the output transistors you can probably go to MJ21194 which has lower absolute collector current but probably higher safe operating area, plus they are cheaper. If you're just going to replace a few output transistor, the old dies don't track over temperature like the new ones, you can't mix them. Make sure 1/2 of them arent PNP transistors, those are MJ21195 or something.
If you have transistors in parallel (probably do with a 2100 watt rating) they need to match. Andrew has said Vce match is more important than gain match, but I don't have a curve tracer, so I put a 200 ohm resistor between Collector and base and a twenty ohm between collector and a 17 VDC battery charger, and measured Vce. Of the ones I bought 10 were the same and two were different, so I used the two as drivers, not as output transistors.
Use a load in the AC when testing, many things before the OT's can be wrong and collapsing the rail voltage can be useful for not exploding new output transistors. A light bulb didn't cut it on my 1300 watt amp, the 15v supply wouldn't make it to run the op amps, so I used a 1500 watt room heater in the AC line. If something was wrong the wires would plink instead of blowing output transistors. Use output resistors instead of speakers for a while, they don't mind DC. I bought mine at Apexelectronic in Sun Valley, 5 ohms 200 w $5, I used 4. I put a cheap car radio speaker in parallel with 5 of the ohms, behind a >1000 uf blocking capacitor (two electrolytics back to back) so I could hear what was going on. For a long time it would work a while then pop and sound bad, until I found the bad solder joint on the input of the op amp of one channel. I found little transistors, diodes, resistors, op amps, and even ceramic caps blown up or opened up by rampant rail voltage, so don't be surprised if you need more than OT's and emmitter resistors. I even had a few traces melted off the PWB. I take it the rail voltage is above 50? Mine was +-85. No rings or other jewelry, use one hand at a time measuring, wear safety glasses in case a die blows out the top of a package.
When done measure idle current of the OT's, should be 20-40 ma for TO3 packages. Also ensure DC out on speakers is <200 mv. See http://www.diyaudio.com/forums/solid-state/136261-vintage-amplifier-repair-upgrade-manual.html
I don't have a scope, or one that works anyway. After the DC was okay, I traced my AC problems with the speakers, and then finally with a VOM with a 2 VAC scale, a blocking capacitors, and a transistor radio on the input. If you tune to to a rock station you can see the music pulsing on the VOM (not DVM) in time with the beat.
If you have transistors in parallel (probably do with a 2100 watt rating) they need to match. Andrew has said Vce match is more important than gain match, but I don't have a curve tracer, so I put a 200 ohm resistor between Collector and base and a twenty ohm between collector and a 17 VDC battery charger, and measured Vce. Of the ones I bought 10 were the same and two were different, so I used the two as drivers, not as output transistors.
Use a load in the AC when testing, many things before the OT's can be wrong and collapsing the rail voltage can be useful for not exploding new output transistors. A light bulb didn't cut it on my 1300 watt amp, the 15v supply wouldn't make it to run the op amps, so I used a 1500 watt room heater in the AC line. If something was wrong the wires would plink instead of blowing output transistors. Use output resistors instead of speakers for a while, they don't mind DC. I bought mine at Apexelectronic in Sun Valley, 5 ohms 200 w $5, I used 4. I put a cheap car radio speaker in parallel with 5 of the ohms, behind a >1000 uf blocking capacitor (two electrolytics back to back) so I could hear what was going on. For a long time it would work a while then pop and sound bad, until I found the bad solder joint on the input of the op amp of one channel. I found little transistors, diodes, resistors, op amps, and even ceramic caps blown up or opened up by rampant rail voltage, so don't be surprised if you need more than OT's and emmitter resistors. I even had a few traces melted off the PWB. I take it the rail voltage is above 50? Mine was +-85. No rings or other jewelry, use one hand at a time measuring, wear safety glasses in case a die blows out the top of a package.
When done measure idle current of the OT's, should be 20-40 ma for TO3 packages. Also ensure DC out on speakers is <200 mv. See http://www.diyaudio.com/forums/solid-state/136261-vintage-amplifier-repair-upgrade-manual.html
I don't have a scope, or one that works anyway. After the DC was okay, I traced my AC problems with the speakers, and then finally with a VOM with a 2 VAC scale, a blocking capacitors, and a transistor radio on the input. If you tune to to a rock station you can see the music pulsing on the VOM (not DVM) in time with the beat.
Last edited:
Indianajo, thanks for your response. There are quite a few posts on this site about the MC2100 amp and the ones that go into replacing the output transistors suggest using the MJ15003G. I could be just over thinking it but am perplexed that the part numbers on the Motorola op transistors all start with 32-0070 and then 4 of them end in 7526, 4 end in 7452, 3 end in 7539, 1 ends in 7513. According the schematic they should all be PNP and all have the same McIntosh part number. Thanks
MJ15003G is a bit of an overkill. 250W 140V. The rail is just 70V and by the way, the original output transistors are NPN the same as MJ5003G, NOT PNP as you describe.
It is not likely that all of the transistors have failed and just to ensure it is the transistors that are blowing the fuses, check them first. It may be the bridge rec or transformer!
It is not likely that all of the transistors have failed and just to ensure it is the transistors that are blowing the fuses, check them first. It may be the bridge rec or transformer!
I found old OT's that passed the double diode DVM test (at 2 VDC) would still explode after a minute or two of 75 v. I built a "Vceo leakage" tester out of the same 12V battery charger (capacitor peaks it at 17 v), with a 47k resistor in series. Then put 20 ma scale of DVM in series. Short b to e, measure leakage current c-e at 17 v. Some don't leak much, some leak a lot. The ones that leak a lot are the damaged ones. when I did that I was able to salvage a couple of transistor that wouldn't blow up. However since MJ21194 is $1.50 cheaper than the MJ15025 my amp was shipped with, I just replaced all the OT's , used one of the leftover MJ15025 as a driver, and put the other in the parts bin. I had MJ15020-21 as a driver, these were probably cheaper in facttoryquantity in 1994, but aren't now. There is no point IMHO in buying 15020 when a 15025 has just as much gain. Don't know about your different part numbers, check carefully that two of them aren't TO3 drivers that the emitter drives the base of the other five on the same heatsink. The drivers can be lower amp rating, but with ON semi g suffix parts having plenty of gain and being the cheapest, there is no reason not to use them as driver as long as you match up left to right channel. In 1990-94 adequate gain was still pricey.
I find transistor amp failures are mostly transistors due to the shorted output plug problem, followed by electrolytic capacitors (after 20 years). If you really want to test your transformer and rectifiers, a load test with the resistors listed above is good for detecting shorted transformer turns. Some transformers will put out voltage okay until you start drawing some actual current, then the voltage collapses. The rectifier bridge, the dvm diode test should be enough IMHO.
I find transistor amp failures are mostly transistors due to the shorted output plug problem, followed by electrolytic capacitors (after 20 years). If you really want to test your transformer and rectifiers, a load test with the resistors listed above is good for detecting shorted transformer turns. Some transformers will put out voltage okay until you start drawing some actual current, then the voltage collapses. The rectifier bridge, the dvm diode test should be enough IMHO.
Last edited:
The 74XX and 75XX numbers are date codes. Call McIntosh, ask for parts dept. and they will be able to tell what generic transistors to use. Usually the drivers were a bit different than the outputs.
Craig
Craig
- Status
- Not open for further replies.