Re: Results..Thank you
Marc,
Bias it to about 10mV across the TP's then run the amp for a while and re-test. If you want you can crank up the bias till the heatsinks touch 40C then crank no more.
If your speakers are demanding, I'd not crank it up more than 30C at idle. Stock 555II's idle at about 26C depending on ambient.
cheers!
mpmarino said:
Marc,
Bias it to about 10mV across the TP's then run the amp for a while and re-test. If you want you can crank up the bias till the heatsinks touch 40C then crank no more.
If your speakers are demanding, I'd not crank it up more than 30C at idle. Stock 555II's idle at about 26C depending on ambient.
cheers!
Hi Marc,
I have posted the factory proceedure earlier in this thread. Follow it. This is more accurate than measuring the heatsink temperatue. If you want to run at a higher bias, don't go too far, the amp normally runs warm. You may get into a situation where, when the ambient temp goes higher, and you push the amp, the amp loses control of the bias current. As a rule of thumb, do not run the junctions of bipolar transistors any hotter than you have to.
As K-amps mentioned in the last post, The 2S numbers are real Toshibas and the house numbers are Motorola. MJ15024 / MJ15025's are the transistors I rebuild these with also.
-Chris
I have posted the factory proceedure earlier in this thread. Follow it. This is more accurate than measuring the heatsink temperatue. If you want to run at a higher bias, don't go too far, the amp normally runs warm. You may get into a situation where, when the ambient temp goes higher, and you push the amp, the amp loses control of the bias current. As a rule of thumb, do not run the junctions of bipolar transistors any hotter than you have to.
As K-amps mentioned in the last post, The 2S numbers are real Toshibas and the house numbers are Motorola. MJ15024 / MJ15025's are the transistors I rebuild these with also.
-Chris
I took an educated 'guess' on using those as well..Superb results.
Thanks for your reply as well, Kamps.
One more thing, the MJ21193/4/5/6 are closer in hFE to the original Toshiba 424/554's than the slower Mj15024/25.
If you need to re-build the 555's use the MJ211XX not the MJ15024/25 unless ofcourse you have many 24/25's at hand.
The 211XX's sound warmer and less harsh.
If you need to re-build the 555's use the MJ211XX not the MJ15024/25 unless ofcourse you have many 24/25's at hand.
The 211XX's sound warmer and less harsh.
K-amps,
Just exactly how fast does an output need to be in this application?? How the devil can they affect the sound that dramatically, enclosed in the feedback loop as they are?
I won't rule out a signal difference, but compared to the other variables at play, I don't think this difference could be heard. I am not sure that a measurement could be made either, given the HF pole of the amp.
BTW, 2SD424's are rated at 5 MHz hFE 40~140, MJ15024's at around 5 MHz GBP hFE 60 (typical) and MJ21194's have a similar GBP rating and hFE spec'd from 25~75. All in all, the two Motorola devices are very similar on these specs. I'd happily use either set.
-Chris
Just exactly how fast does an output need to be in this application?? How the devil can they affect the sound that dramatically, enclosed in the feedback loop as they are?
I won't rule out a signal difference, but compared to the other variables at play, I don't think this difference could be heard. I am not sure that a measurement could be made either, given the HF pole of the amp.
BTW, 2SD424's are rated at 5 MHz hFE 40~140, MJ15024's at around 5 MHz GBP hFE 60 (typical) and MJ21194's have a similar GBP rating and hFE spec'd from 25~75. All in all, the two Motorola devices are very similar on these specs. I'd happily use either set.
-Chris
anatech said:
Trust no one but your ears Chris, For a Tin eared person like me, it makes a difference. ;-)
The 24/25's in the Adcom running a 4 ohm load have hFE more in the 25 region than 60, the 2119X are higher. Actually at Ic of 8A, the 24/25 is more like 15 and the 21195/96 is more like 25.
I forgot where but I saw hfe curves of the 1302 vs. 2119X vs. the 15025 vs. the 3055. (probably in Self's book)
The 2119X was consistently higher in gain, i.e. puts less strain on the drivers where most of the beta droop related distortions occur. Additionally the 24/25 have a cob of 600pF, thats 100 more than the 2119X with cob of 500, meaning a higher loading on the drivers. With the way the 555 is designed, it will run better with the 2119X than the 15024/25 even though both will operate without issues.
Another difference with the 25's and the 2119X's is that the 24/25 are constructed using MOT's powerbase technology while the 2119X's use the same process as the 2sd424/554 i.e. perforated emitter technology.
Not sure if that make a sonic difference but it sure makes me feel better. 😉
The final factor however was a listening test... the 24/25's do not sound as lush and detailed as the TOS or the 2119X's granted they are much more robust than the Toshiba's.
The 24/25 are cheaper though.... but then I can get the 2119X in metal can for less than $2 each...
😀
Hi K-amps,
I'll agree with you on the perforated "multi-emitter". It is much more linear and is similar to the popular Japanese outputs. Much flatter hFE vs current characteristic. Since I am designing an amp at the moment, I may have a chance to actually compare the two side by side. I can't remember hearing a difference in the past.
-Chris
I'll agree with you on the perforated "multi-emitter". It is much more linear and is similar to the popular Japanese outputs. Much flatter hFE vs current characteristic. Since I am designing an amp at the moment, I may have a chance to actually compare the two side by side. I can't remember hearing a difference in the past.
-Chris
Great info here, I've been lurking in the background trying to make a decision on whether to send in my 555 to Adcom or attempt a fix myself. I do have a Q, though, and would appreciate any help.
Amp immediately blows fuse on one bank of outputs, suspecting blown output transistors (output leads from board showed zero resistance between leads, other 3 boards don't) I disassembled. Much to my surprise the 4 2SB554 devices test ok. However, Q11 which is labeled 10123 is shorted between leads. I am assuming here that this is the same component shown on schematic as Q11, 2SA1011 or 2SB595. Of the four output boards, only two of them have a small cap in parallel with this component though the schematic shows them all having a cap.
Is this a critical component that needs a direct replacement or can I use a generic replacement? Kinda hate to mail order a .67 cent component if I can pick something up locally. I'm also wondering if this device being shorted might be symptomatic of a problem in the front end board. Hopefully not cause I'll be one lost cause trying to fix that!
I'd appreciate you gurus' opinion on this. Maybe I'll have tunes again someday.
Regards, Chuck
Amp immediately blows fuse on one bank of outputs, suspecting blown output transistors (output leads from board showed zero resistance between leads, other 3 boards don't) I disassembled. Much to my surprise the 4 2SB554 devices test ok. However, Q11 which is labeled 10123 is shorted between leads. I am assuming here that this is the same component shown on schematic as Q11, 2SA1011 or 2SB595. Of the four output boards, only two of them have a small cap in parallel with this component though the schematic shows them all having a cap.
Is this a critical component that needs a direct replacement or can I use a generic replacement? Kinda hate to mail order a .67 cent component if I can pick something up locally. I'm also wondering if this device being shorted might be symptomatic of a problem in the front end board. Hopefully not cause I'll be one lost cause trying to fix that!
I'd appreciate you gurus' opinion on this. Maybe I'll have tunes again someday.
Regards, Chuck
interesting test...
I just finished a repair of 2 identical amps, a pair of old adcom gfa1's. I had enough 424/554s to refit one with original transistors, the other got 15003/4's.
They measure identically, to the limits of my equipment, but I haven't taken the time to listen yet.
The transistors are socketed and I can change out the entire set in about 10 mins, so I may just have to buy a few more transistors for a soundoff...
Is there anyone out there in or around santa cruz, ca that would be interested in providing more ears? I am not sure I am a discriminating enough listener...
Stuart
I just finished a repair of 2 identical amps, a pair of old adcom gfa1's. I had enough 424/554s to refit one with original transistors, the other got 15003/4's.
They measure identically, to the limits of my equipment, but I haven't taken the time to listen yet.
The transistors are socketed and I can change out the entire set in about 10 mins, so I may just have to buy a few more transistors for a soundoff...
Is there anyone out there in or around santa cruz, ca that would be interested in providing more ears? I am not sure I am a discriminating enough listener...
Stuart
Hi Chuck,
2SA1011. The 2SB595 does not have a high enough voltage breakdown (100V). Replace with real Japanese, not ECG / NTE or JAN numbers. They are different.
The best why to service check transistors is to measure gain and leakage. Heathkit made a great meter (IT-18). They made a better one in the IT-121, but this one is too slow to use by comparison. Your meter test can miss small leakage currents that is an indicator of transistor failure (short in near future). Test all transistors at least one stage behind the last defective one you find. Replace all forward. You can carefully test your outputs, a good service shop would replace that channel. Do not forget to test all resistors. The ones between emitters tend to open.
-Chris
2SA1011. The 2SB595 does not have a high enough voltage breakdown (100V). Replace with real Japanese, not ECG / NTE or JAN numbers. They are different.
The best why to service check transistors is to measure gain and leakage. Heathkit made a great meter (IT-18). They made a better one in the IT-121, but this one is too slow to use by comparison. Your meter test can miss small leakage currents that is an indicator of transistor failure (short in near future). Test all transistors at least one stage behind the last defective one you find. Replace all forward. You can carefully test your outputs, a good service shop would replace that channel. Do not forget to test all resistors. The ones between emitters tend to open.
-Chris
anatech said:
Yup, If I remember correctly Nelson used the MJE15030/31 pair as drivers for the original 555 design. Also you can use the 2sd1047 / 2sb817 pair Adcom used in the 555 mk. II. and several Japanese subs.
Hi K-amps,
The Japanese units have a higher FT and flatter gain vs current curve. Yes, the MJ's will work. This may not be that important in the long run for Chuck.
Chuck,
Do not buy the cheapest transistors. Always buy from a known good source and watch for re-marks. I'm seeing more and more of these from discount suppliers. Work carefully and take your time, that means wait for a backorder rather than just sticking something in to see if it works.
The Japanese units have a higher FT and flatter gain vs current curve. Yes, the MJ's will work. This may not be that important in the long run for Chuck.
Chuck,
Do not buy the cheapest transistors. Always buy from a known good source and watch for re-marks. I'm seeing more and more of these from discount suppliers. Work carefully and take your time, that means wait for a backorder rather than just sticking something in to see if it works.
Chris and K-Amps
Thank you so very much for your advice and suggestions. I'll source the best components as you suggested. I'm also going to look into a transistor tester though I am considering changing out all output transistors to one of the suggested upgrades.
I called Adcom today and spoke with their parts person. After discussing 555 repairs with her I'm convinced that this is the right track to take considering cost/value of this old amp. Old it may be but it's provided joyful hours and it's far too impressive a piece of equipment to hit the bin. Interestingly, she stated that Adcom would not field match the output transistors if I purchase from them. I didn't ask if this is a service they provide when they are doing the repairs.
Thanks again for your help.
Chuck
Thank you so very much for your advice and suggestions. I'll source the best components as you suggested. I'm also going to look into a transistor tester though I am considering changing out all output transistors to one of the suggested upgrades.
I called Adcom today and spoke with their parts person. After discussing 555 repairs with her I'm convinced that this is the right track to take considering cost/value of this old amp. Old it may be but it's provided joyful hours and it's far too impressive a piece of equipment to hit the bin. Interestingly, she stated that Adcom would not field match the output transistors if I purchase from them. I didn't ask if this is a service they provide when they are doing the repairs.
Thanks again for your help.
Chuck
Adcom originally used the 2SA1011, they frequently broke when driving low impedance loads. The 2SB595 was their fix, they SELECTED this part.
"The Japanese units have a higher FT and flatter gain vs current curve."
Considering that the gain curve for a MJE150XX part looks like it was drawn with a ruler it makes a good replacement in this application.
The 2SA1011 was intended to be a predriver for a 100W amplifier, not the driver for a 200W amplifier, the current gain drops like a rock above 200mA or so.
"The Japanese units have a higher FT and flatter gain vs current curve."
Considering that the gain curve for a MJE150XX part looks like it was drawn with a ruler it makes a good replacement in this application.
The 2SA1011 was intended to be a predriver for a 100W amplifier, not the driver for a 200W amplifier, the current gain drops like a rock above 200mA or so.
Hi Chuck,
Good to know Adcom no longer sells transistors matched. I haven't talked to them in a few years. When I was warranty, it was expected from us to match the parts on a warranty repair. Out of warranty was expected to be at the same levels of service.
djk,
Adcom amps were frequently found driving silly loads, or bridged 4 ohm loads (equally silly). There were changes to make the amp stand up to this non-rated situation. 2SB595 / 2SD525's would blow on clipped 8 ohm loads (or open). Big surprise. You will see amps with both types from the factory. 2SA1011 / 2SC2344 are safer when the amp is in a normal situation. 2SD525 / 2SB595's will blow up for sure at higher voltage swings, don't even think of using these. MJ15003 / MJ15003 are TO-3 outputs with a 2MHz GBP. MJ15024 / MJ15025 are superiour here with a lower 16A Ic and 5MHZ GBP. MJ15030 and 15031 would replace the drivers.
-Chris
Good to know Adcom no longer sells transistors matched. I haven't talked to them in a few years. When I was warranty, it was expected from us to match the parts on a warranty repair. Out of warranty was expected to be at the same levels of service.
djk,
Adcom amps were frequently found driving silly loads, or bridged 4 ohm loads (equally silly). There were changes to make the amp stand up to this non-rated situation. 2SB595 / 2SD525's would blow on clipped 8 ohm loads (or open). Big surprise. You will see amps with both types from the factory. 2SA1011 / 2SC2344 are safer when the amp is in a normal situation. 2SD525 / 2SB595's will blow up for sure at higher voltage swings, don't even think of using these. MJ15003 / MJ15003 are TO-3 outputs with a 2MHz GBP. MJ15024 / MJ15025 are superiour here with a lower 16A Ic and 5MHZ GBP. MJ15030 and 15031 would replace the drivers.
-Chris
anatech said:
Chris... at the 164vdc rails, would'nt we be over the MJ15003/4 's vce rating of 140v 😉
I know I have successfully used MOT MJ15003/4 with 180vdc rails (90 +90) with no issues. Not that I recommend doing it Chuck. 😀
Hi K-amps,
Absolutely. I think with a low impedance load, someone hopes for no clipping. They pray for no clipping with no load.
I don't normally use these anyhow. You would probably get away with this setup unless you got one transistor with a lower Vceo than the others. Luck of the draw here since there is negative headroom.
-Chris
Absolutely. I think with a low impedance load, someone hopes for no clipping. They pray for no clipping with no load.
I don't normally use these anyhow. You would probably get away with this setup unless you got one transistor with a lower Vceo than the others. Luck of the draw here since there is negative headroom.
-Chris
I think you may have missed the point about Adcom 595/525s, they were selected on a curve tracer for Vceo, they are not stock items from a distributor.
The 1011/2344 would blow whenever the B± fuses blew, this would most often happen driving low impedance, low efficency speakers like Infinity.
The outputs were usually OK.
The SOA stinks on the MJ15003/04 above 50V, MJ15024/25 are OK to 80V, MJ15011/12 have a higher safe area at 100V that the MJ15024/25, The MJ21193/94 are now the 100V king, but the MJ21195/96 have the best at 80V.
Actually, the all time SOA king at 100V was the 2N6259 (SOA of 250W at 100V), used in the Altec 9440 and BGW 1000. The MJ2119X series is the closest currently made.
The 1011/2344 would blow whenever the B± fuses blew, this would most often happen driving low impedance, low efficency speakers like Infinity.
The outputs were usually OK.
The SOA stinks on the MJ15003/04 above 50V, MJ15024/25 are OK to 80V, MJ15011/12 have a higher safe area at 100V that the MJ15024/25, The MJ21193/94 are now the 100V king, but the MJ21195/96 have the best at 80V.
Actually, the all time SOA king at 100V was the 2N6259 (SOA of 250W at 100V), used in the Altec 9440 and BGW 1000. The MJ2119X series is the closest currently made.
O.K. I think I got it...
I'm going to use the MJE15030/31 part for the drivers. Will replace em all.
I'm chasing a Heathkit transistor tester, when I get it I'll check the output transistors, right now they seem to be ok but I wonder how in spec they are after 20 years or so.
The .82 , 2 watt resistors associated with output transistors, are these metal film, or ceramic or does it matter? I intend to check these with a sensitive vom at work but want to get a few in with my digikey order. Can you think of any other likely suspects in the electronic lineup?
I'm pretty excited about the prospects for fixing this amp, ordered a service manual and now have lustful dreams of stacks of overhauled 555's driving whatever speakers I want to throw at em. Once again guys thanks for your help.
Chuck
I'm going to use the MJE15030/31 part for the drivers. Will replace em all.
I'm chasing a Heathkit transistor tester, when I get it I'll check the output transistors, right now they seem to be ok but I wonder how in spec they are after 20 years or so.
The .82 , 2 watt resistors associated with output transistors, are these metal film, or ceramic or does it matter? I intend to check these with a sensitive vom at work but want to get a few in with my digikey order. Can you think of any other likely suspects in the electronic lineup?
I'm pretty excited about the prospects for fixing this amp, ordered a service manual and now have lustful dreams of stacks of overhauled 555's driving whatever speakers I want to throw at em. Once again guys thanks for your help.
Chuck
- Status
- Not open for further replies.