I have an older (circa 1988) Bryston 3B that has blown two output transistors on one channel.
They seem to have house part numbers on them, BR6521 and BR6522. They are of Motorola manufacture. The amp appears to have been repaired at some point as there are two newer transistors. They also appear to be house part numbers, SJ6521 and SJ6522, also from Motorola. The drivers are SJE1490 and SJE1491, again Motorola parts.
Does anyone know with reasonable certainty if these are simply selected standard devices and what the standard equivalents would be? Newer Versions of the 3B use MJL21193 and MJL21194 for outputs (yes, a different case style) and MJE15030 and MJE15031 for the drivers.
Any thoughts on using MJ21193/MJ21194, MJ21195/MJ21196, or MJ15003/MJ15004 pairs? I'll assume the selection criteria will be good match in hFE and Vbe. The MJ2119x have a little more SOA at higher voltages and a higher fT at the possible expense of hFE. The MJ1500x have similar SOA at the amplifier's operating voltage, but loose a little at higher voltages. Also fT is lower, but hFE could potentially be higher. Or should I simply see if Bryston will sell spares?
It seems there is no collateral damage, but if anyone is well versed in servicing these then I'm open to suggestions on what to look at before putting it back into service.
This was a freebie, but I'll spend some money on it if it is worth pursuing.
They seem to have house part numbers on them, BR6521 and BR6522. They are of Motorola manufacture. The amp appears to have been repaired at some point as there are two newer transistors. They also appear to be house part numbers, SJ6521 and SJ6522, also from Motorola. The drivers are SJE1490 and SJE1491, again Motorola parts.
Does anyone know with reasonable certainty if these are simply selected standard devices and what the standard equivalents would be? Newer Versions of the 3B use MJL21193 and MJL21194 for outputs (yes, a different case style) and MJE15030 and MJE15031 for the drivers.
Any thoughts on using MJ21193/MJ21194, MJ21195/MJ21196, or MJ15003/MJ15004 pairs? I'll assume the selection criteria will be good match in hFE and Vbe. The MJ2119x have a little more SOA at higher voltages and a higher fT at the possible expense of hFE. The MJ1500x have similar SOA at the amplifier's operating voltage, but loose a little at higher voltages. Also fT is lower, but hFE could potentially be higher. Or should I simply see if Bryston will sell spares?
It seems there is no collateral damage, but if anyone is well versed in servicing these then I'm open to suggestions on what to look at before putting it back into service.
This was a freebie, but I'll spend some money on it if it is worth pursuing.
Read this thread:
Bryston 3B repair.. What should I do? - Gearslutz.com
Despite the naming in the link it is about a Bryston 3B.
The link is a US based situation...Try to call Bryston Canada and see if their policy and pricing is similar.
Bryston 3B repair.. What should I do? - Gearslutz.com
Despite the naming in the link it is about a Bryston 3B.
The link is a US based situation...Try to call Bryston Canada and see if their policy and pricing is similar.
If you don't want to pursue repair, let me know...I might be interested in a drive to Paris for the right price.
Darn, just found this link:
D 'n A Electronics Specifications | Bryston 3B Specs | Bryston 3B Repair | Bryston Repair | Bryston Stereo Repair | Vintage Stereo Repair
"Bryston 3B Restoration:
We restore 3B power amplifiers."
Let me know either way.
Darn, just found this link:
D 'n A Electronics Specifications | Bryston 3B Specs | Bryston 3B Repair | Bryston Repair | Bryston Stereo Repair | Vintage Stereo Repair
"Bryston 3B Restoration:
We restore 3B power amplifiers."
Let me know either way.
I would really like to have a listen to this unit before I decide wether or not it will have a permanent home here or not. I have it stripped down and have it all cleaned up, I just need to fix the bad channel. These don't look like they are exceptionally complicated and I don't think it will be tough to set right. I comes down to parts selection / sourcing at the moment.
I have some other 'rescued' old pro amps, one all cleaned up and working (Yamaha PD2500) and another I also just got (HH V800) that doesn't seem to power up properly (I think the voltage selector switch is toast). That Yamaha is actually nice sounding, though the fan cooling is a bummer in a quiet setting. The HH has yet to sing, we'll see. It is a massive over-built british beast with a wiring loom that looks like a nightmare and also fan cooled. Comparatively the Bryston is nicely made and and convection cooled. I'm looking forward to hearing it.
I have some other 'rescued' old pro amps, one all cleaned up and working (Yamaha PD2500) and another I also just got (HH V800) that doesn't seem to power up properly (I think the voltage selector switch is toast). That Yamaha is actually nice sounding, though the fan cooling is a bummer in a quiet setting. The HH has yet to sing, we'll see. It is a massive over-built british beast with a wiring loom that looks like a nightmare and also fan cooled. Comparatively the Bryston is nicely made and and convection cooled. I'm looking forward to hearing it.
The company that is retiring some of their gear (and where my existing pieces were sourced from) apparently has more 3B and 4B available, but since they apparently work they aren't free. Something like $200 per 3B and $500 per 4B. I've no idea if that's reasonable for ex-sound reinforcement equipment. Any thoughts?
If they are anything like the equipment I have already gotten from them then they will be well travelled with a smattering of abuse and inexpert repair / modification, oh, and dirty inside. I have stripped every piece I have ever gotten from them to the base plate and built it back up. But then, I also wait for the scraps to fall off the table, so to speak. Sounds like they want premium dollar.
No one has any information about replacing the 3B's transistors with industry standards?
I got a message back from Bryston to say they will sell the parts at $15 CAD plus tax and shipping. Any idea how many standard parts like the MJ21195 / MJ21196 would have to be ordered to achieve a suitable match?
It may be worth it to pay the premium to buy the OEM devices.
I got a message back from Bryston to say they will sell the parts at $15 CAD plus tax and shipping. Any idea how many standard parts like the MJ21195 / MJ21196 would have to be ordered to achieve a suitable match?
It may be worth it to pay the premium to buy the OEM devices.
Thanks for that.
I actually had some of the MJ15024/MJ15025 and even more of the MJ15003/MJ15004 already on hand. I couldn't resist, so I dropped in a set of the MJ15003/MJ15004 just to see if the thing would power up. Seemed to come up just fine (via a bulb limiter of course). Bias on both sets of TP adjusted and seemed pretty even, there was even lower DC offset than the original good channel (5mV new side, 13mV original side).
Anyone know what the bias value is supposed to be for this vintage 3B (Bryston Spec)? I measured the all original channel and it was about 6mV / 25mA but that seems a little low to me. Perhaps it isn't considering what this amp was built to do.
Seems to be accepted that the best performance comes when the is about 26mV across the emitter resistors. This would be a little over 100mA per device. Quiescent dissipation would be close to 6W per transistor, 24W per side. That seems a little high to me. If no one knows I'll see how stable it is at about 50mA per device, 12mV at the test points.
Suggestions welcome.
I actually had some of the MJ15024/MJ15025 and even more of the MJ15003/MJ15004 already on hand. I couldn't resist, so I dropped in a set of the MJ15003/MJ15004 just to see if the thing would power up. Seemed to come up just fine (via a bulb limiter of course). Bias on both sets of TP adjusted and seemed pretty even, there was even lower DC offset than the original good channel (5mV new side, 13mV original side).
Anyone know what the bias value is supposed to be for this vintage 3B (Bryston Spec)? I measured the all original channel and it was about 6mV / 25mA but that seems a little low to me. Perhaps it isn't considering what this amp was built to do.
Seems to be accepted that the best performance comes when the is about 26mV across the emitter resistors. This would be a little over 100mA per device. Quiescent dissipation would be close to 6W per transistor, 24W per side. That seems a little high to me. If no one knows I'll see how stable it is at about 50mA per device, 12mV at the test points.
Suggestions welcome.
As an aside, considering the pairs I have on hand, how much difference is it likely to make which one is used? The amp operates on +/-55V rails.
MJ15003/MJ15004
140V
20A
250W
SOA@50V=5A
fT=2MHz
Cob=1000pF
hFE=25-150
MJ15024/MJ15025
250V
15A
250W
SOA@50V=5A
fT=4MHz
Cob=500pF
hFE=15-60
Both have the same junction to case thermal resistance and derating factor.
I'm wondering if the possible higher hFE might be worth the trade off in speed? Thoughts...
MJ15003/MJ15004
140V
20A
250W
SOA@50V=5A
fT=2MHz
Cob=1000pF
hFE=25-150
MJ15024/MJ15025
250V
15A
250W
SOA@50V=5A
fT=4MHz
Cob=500pF
hFE=15-60
Both have the same junction to case thermal resistance and derating factor.
I'm wondering if the possible higher hFE might be worth the trade off in speed? Thoughts...
My own personal hunch would be to use the highest reliability choice, especially if these are the original 3B / 4B type units from the 1970's+.
Bryston built these amps to be rugged low-failure units for professional applications. They were initially surprised to learn that people were buying them for HiFi sound quality purposes. They did not really make design choices for sound quality specifically until the next series, once they realized a market existed outside their original intended one.
As far as my own personal opinion and experiences with listening to (not repairing, which I've never attempted) the early Bryston amps, for SQ purposes I would rank them:
Best: 2B
Next Best: 4B
The 2B and 4B are reasonably close to each other; if you need the power the 4B is acceptable over the 2B.
Finally (and noticeable step down): 3B
The subsequent series amps all sound better, But note that the 2B was not always improved in step, yet held it's own.
Bryston built these amps to be rugged low-failure units for professional applications. They were initially surprised to learn that people were buying them for HiFi sound quality purposes. They did not really make design choices for sound quality specifically until the next series, once they realized a market existed outside their original intended one.
As far as my own personal opinion and experiences with listening to (not repairing, which I've never attempted) the early Bryston amps, for SQ purposes I would rank them:
Best: 2B
Next Best: 4B
The 2B and 4B are reasonably close to each other; if you need the power the 4B is acceptable over the 2B.
Finally (and noticeable step down): 3B
The subsequent series amps all sound better, But note that the 2B was not always improved in step, yet held it's own.
Last edited:
In looking at the graphs they actually look pretty close to me (ON Semi data sheets) at the 5A mark with hFE=40ish for each device type. I picked the MJ15003/MJ15004 because I had twice as many of them, so in the event they ended up being sacrificial I wasn't out of my MJ15024/MJ15025 pairs.
The outputs and drivers are in sockets, so I can actually try both.
And Rock-n'-Roll it does!
I actually ended up fitting a second set of the MJ15003/MJ15004 into the good original channel too. Reason was the bias setting was more even on the new side than the original and I wanted to see if that was just luck or not. Once satisfied, off the limiter it came and time for bias set-up. Gave it a good long time to stabilize and got the bias set up to the specification graciously provided by Bryston, being 10mV BTW. The bias was actually more evenly split between the devices than with the original outputs, go figure. So, there goes the idea of device matching / selection?
Played it for some time into some crappy speakers to give it a chance to operate into a load and convince me all was well. Moved it up to the living room and hooked it up to my main system. Played a variety of music and found this amp is quite pleasant with excellent extension into the bass region and smooth sounding treble.
I'm thinking I might have to place this into my system for a long term listen. Going back to back with my Yamaha AX-570 (yeah, yeah, I know, not great) it was much more pleasant. The Yamaha has much sharper treble, to the point of being comparatively fatiguing and only now shows how anemic is is down low.
Despite the thread originally being about the output transistors, I'd like to swing it towards a more holistic refurbishing. The connectors, gain controls, and switches are all quiet and in otherwise good condition so I have no worries there. The only thing really left to do is replace the few electrolytic capacitors in it, aside from the main filter caps. I'm thinking of these for the PSU. Not sure about the three (per channel) 470uF 16V aluminum electrolytic capacitors, two are just for some filtering / decoupling, the other is the DC blocking cap in the feedback path. There are also two tantalum capacitors per channel, decoupling for the zener diodes feeding the front end. Are these usually part of refurbishment too? Any value to retrofitting some nice film / mica caps for the small value units?
Only issue now is that my cabinet is only 17 1/4" wide and this is a rack mount piece at 19" wide. Be a shame to have to physically cut it, I'll have to see about alternative placement.
I actually ended up fitting a second set of the MJ15003/MJ15004 into the good original channel too. Reason was the bias setting was more even on the new side than the original and I wanted to see if that was just luck or not. Once satisfied, off the limiter it came and time for bias set-up. Gave it a good long time to stabilize and got the bias set up to the specification graciously provided by Bryston, being 10mV BTW. The bias was actually more evenly split between the devices than with the original outputs, go figure. So, there goes the idea of device matching / selection?
Played it for some time into some crappy speakers to give it a chance to operate into a load and convince me all was well. Moved it up to the living room and hooked it up to my main system. Played a variety of music and found this amp is quite pleasant with excellent extension into the bass region and smooth sounding treble.
I'm thinking I might have to place this into my system for a long term listen. Going back to back with my Yamaha AX-570 (yeah, yeah, I know, not great) it was much more pleasant. The Yamaha has much sharper treble, to the point of being comparatively fatiguing and only now shows how anemic is is down low.
Despite the thread originally being about the output transistors, I'd like to swing it towards a more holistic refurbishing. The connectors, gain controls, and switches are all quiet and in otherwise good condition so I have no worries there. The only thing really left to do is replace the few electrolytic capacitors in it, aside from the main filter caps. I'm thinking of these for the PSU. Not sure about the three (per channel) 470uF 16V aluminum electrolytic capacitors, two are just for some filtering / decoupling, the other is the DC blocking cap in the feedback path. There are also two tantalum capacitors per channel, decoupling for the zener diodes feeding the front end. Are these usually part of refurbishment too? Any value to retrofitting some nice film / mica caps for the small value units?
Only issue now is that my cabinet is only 17 1/4" wide and this is a rack mount piece at 19" wide. Be a shame to have to physically cut it, I'll have to see about alternative placement.
Look at the tantalums. If one lead is starting to turn black it's time to replace them, they're leaking.
Replace the cap in the feedback loop with as high a quality as what you can find. I tack on a film bypass to improve HF detail, YMMV.
The main filter caps on those do go bad, hard to find the right size and value. I tack on a small electrolytic bypass to tighten up the bass, YMMV.
With the bypass caps installed it sounds more like the Leach, a good thing in my experience.
Replace the cap in the feedback loop with as high a quality as what you can find. I tack on a film bypass to improve HF detail, YMMV.
The main filter caps on those do go bad, hard to find the right size and value. I tack on a small electrolytic bypass to tighten up the bass, YMMV.
With the bypass caps installed it sounds more like the Leach, a good thing in my experience.
Hi Guys
MJ15024/25 are what the original Bryston output devices were based on. These have twice the Ft of the MJ15003/4.
Possibly of more importance is the voltage rating: 250V for the 24/25 vs 140V for the 3/4. The outputs can see the full width of the supply, which is 180V in newer versions, maybe 160V in older.
Leach amps are quite good and benefit from matched devices. Even when fitted, THD is still 10-50 times as high a Bryston properly set up.
The new amps use TO-3P case outputs instead of TO-3. You could certainly use the flat packs on the old heatsinks but the old design was predicated on the extra 25-50C capability of the all-metal case.
Have fun
Kevin O'Connor
Bryston offers a 20 year warranty and is quite reasonable and generous with parts for older units. It is always worth asking them for advice.
MJ15024/25 are what the original Bryston output devices were based on. These have twice the Ft of the MJ15003/4.
Possibly of more importance is the voltage rating: 250V for the 24/25 vs 140V for the 3/4. The outputs can see the full width of the supply, which is 180V in newer versions, maybe 160V in older.
Leach amps are quite good and benefit from matched devices. Even when fitted, THD is still 10-50 times as high a Bryston properly set up.
The new amps use TO-3P case outputs instead of TO-3. You could certainly use the flat packs on the old heatsinks but the old design was predicated on the extra 25-50C capability of the all-metal case.
Have fun
Kevin O'Connor
Bryston offers a 20 year warranty and is quite reasonable and generous with parts for older units. It is always worth asking them for advice.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.