Hi folks,
in a friend's active speakers from the early 90s, work with IRF130 and IRF9130 VMOSFETs on the power amp modules. In one of the modules both VMOSFETs have failed (seemingly due to electric overvoltage). So far as we can find, the parts are not available from reliable sources anymore and other sources seem rather doubious...
Please, anyone, if at all possible, please advise which parts would work drop-in compatible or what we can or have to do to repair the power module! We do actually have the power amp's schematics if needed.
Thanks and Regards,
Winfried
in a friend's active speakers from the early 90s, work with IRF130 and IRF9130 VMOSFETs on the power amp modules. In one of the modules both VMOSFETs have failed (seemingly due to electric overvoltage). So far as we can find, the parts are not available from reliable sources anymore and other sources seem rather doubious...
Please, anyone, if at all possible, please advise which parts would work drop-in compatible or what we can or have to do to repair the power module! We do actually have the power amp's schematics if needed.
Thanks and Regards,
Winfried
1) please post the schematic.
If available, it should always accompany help requests.
If anything, it can help others in the future with the same amp.
2) it will be hard to find TO3 packaged MosFets these days, not made any more for a long time.
In principle, modern plastic package ones such as IRF240/9240 which by the way are way more robust , have dimensions compatible with TO3 metallic, I mean pin separation and one mounting bolt hole.
All the time I do that substitution, bending Gate and Source leads down so they fit original PCB holes (you might need to enlarge them a little), In use a 1.25mm drill because leads are flat but wider, not round, Drain is automatically connected through the mounting bolt as before, sometimes the "missing" bolt is needed because PCB designer used TO3 case as a "bridge" between two tracks, in that case mount a bolt with a terminal there and solder "flying" Drain leg to it.
Plastic cases have two drain terminals: metal tab/body and center leg.
Some plastic cases have fully insulated (plastic covered) mounting holes, in that case use the leg-to-terminal approach and if needed connect the pad on the solder side with a short piece of wire.
In a nutshell: you CAN very well replace a metallic TO3 case transistor with a plastic case one, it only takes a little bit of extra work.
Give the sheer lack of "originals", it´s the best alternative, I do it all the time.
That said, IF you get originals or compatible ones which are not Fake, go for it, of course.
Just curious, are those Montarbo powered speakers?
If available, it should always accompany help requests.
If anything, it can help others in the future with the same amp.
2) it will be hard to find TO3 packaged MosFets these days, not made any more for a long time.
In principle, modern plastic package ones such as IRF240/9240 which by the way are way more robust , have dimensions compatible with TO3 metallic, I mean pin separation and one mounting bolt hole.
All the time I do that substitution, bending Gate and Source leads down so they fit original PCB holes (you might need to enlarge them a little), In use a 1.25mm drill because leads are flat but wider, not round, Drain is automatically connected through the mounting bolt as before, sometimes the "missing" bolt is needed because PCB designer used TO3 case as a "bridge" between two tracks, in that case mount a bolt with a terminal there and solder "flying" Drain leg to it.
Plastic cases have two drain terminals: metal tab/body and center leg.
Some plastic cases have fully insulated (plastic covered) mounting holes, in that case use the leg-to-terminal approach and if needed connect the pad on the solder side with a short piece of wire.
In a nutshell: you CAN very well replace a metallic TO3 case transistor with a plastic case one, it only takes a little bit of extra work.
Give the sheer lack of "originals", it´s the best alternative, I do it all the time.
That said, IF you get originals or compatible ones which are not Fake, go for it, of course.
Just curious, are those Montarbo powered speakers?
Hello JM,
the speakers are german made Baclkes&Müller BM-8 VFET from the early 90s.
Using plastic package PowerTransistors is a good idea to consider and basically a "no-brainer" to build into the applicaiton.
My friend bought "original replacements" and one of them seems to be a fake, so your advice is well appreciated!
So, the real question is/remains which (non-TO-3) VFETs can replace the IRF130/IRF9130? And do I understand correctly that IRF240/IRF9240 should work? I'm not competent enough to judge that myself...
Here's the schematic:
Thanks for all advice!
Winfried
the speakers are german made Baclkes&Müller BM-8 VFET from the early 90s.
Using plastic package PowerTransistors is a good idea to consider and basically a "no-brainer" to build into the applicaiton.
My friend bought "original replacements" and one of them seems to be a fake, so your advice is well appreciated!
So, the real question is/remains which (non-TO-3) VFETs can replace the IRF130/IRF9130? And do I understand correctly that IRF240/IRF9240 should work? I'm not competent enough to judge that myself...
Here's the schematic:
Thanks for all advice!
Winfried
English please.Hi Winfried,
Facing the same problem. BM20 Ebay 4 Power Amps Dead.
Had you found replacement guys?
Are the IRFP140 and IRFP9140 compatible with the IRF130 and IRF9140?
Best regards from Austria and
thank you very much for your answer
Michael
The VN10KM is a high speed RF capable device and characterized for linear operation. Being used in a cascode with an MJE340 as the active device which takes all the voltage. Likely a lot of things will work here (even a good bipolar as you are current source biased so vgs(vbe) is almos immaterial. The VN10KM was pretty nice in it’s day and I’ve used lots of them, but technology has moved on, and I’m sure you could find one equivalent or better if you expand your search beyond “audio” or “switching” types and look into the RF domain. The trick is NOT to oversize it. An IRFP140 would physically “work“ but I’m sure it would degrade the sound.
And if you find those 2N part numbers they will likely be fake, overpriced, or both. Gone are the days of finding old Unitorode and such surplus for like two bucks. Those disappeared when IR did away with all the TO-3s (Which replaced the 2Ns, pretty much industry wide except for military where it takes an act of Congress to substitute and equivalent part). Now if you find them surplus they’re jacked to the stratosphere. Can probably get “something” marked as such on E-bay (along with the $1 2SC5200’s). Stick to the 240/9240 while you can get them, and if you use them a lot keep your eyes peeled for the EOL notice, at least weekly. Unless you’ve already got a plan in place.
I have a customer's Counterpoint SA-12 that has blown output transistors. Looks like IRFP240 and IRFP9240 are the only viable replacements. But, close matching is required to set the bias to the recommended 280mV. (On a prior repair, I found it impossible to set the bias under 460mV because the matching was not close enough.) So, what's the best way to get matched quads of these?
Thanks.
Thanks.
"Bias" could mean the DC offset or the source resistor idle voltage drop, ie idle current??? The DC offset of most amps is a function of the input LTP but if this is that tube driven circuit there is no LTP and the MOS outputs are capacitor coupled from the tubes. The DC offset is set by a resistor divider, which is not difficult to modify for more range. The schematic I found is not symmetric and applies DC bias to the P-channels side (only), a design issue that should be corrected.
Attachments
Oh, the design is symmetric, bias is also symmetric.
I guess you are talking offset correction
But to correct offset you need to apply a small voltage to one of the transistors, the other will stay "one bias drop" away from the other, no matter what.
Here it could be done on the N side or the P side, flip a coin.
There is nothing to correct , it works fine as shown 🙂
Offset network provides ample adjustment range, +/- 8V around center/ground voltage.
Your post is not clear enough.
1) bias and offset have already been confused in this thread, which one are you talking about?
2) where should those "280mV" be measured?
3) given that, we need the schematic.
Sorry, I assumed the schematic had already been posted to this thread. The SA-100 schematic that SteveU posted is almost identical to the SA-12, except for the addition of .22Ω source resistors. Since the SA-12 lacks these, bias current has to measured by breaking the supply circuit. Rather then inserting a DC ammeter as some manufacturers suggest, Counterpoint suggests removing one of the rail fuses and inserting a blown fuse with a 1Ω power resistor soldered across it and measuring the DC voltage across that. It says to set the bias to 280mV (corresponding to 280mA for both devices in that rail). Then, it says to adjust the offset to 0 (plus or minus 20mV). And then to re-check bias and offset after a few minutes.
Now, on a prior repair of a different SA-12, the MOSFETS were not closely matched, and the bias could not be reduced below 460mA. Once I installed a closely-matched quad, I had no problem reducing the bias to 280mA. So, considering the difficulty of sourcing matched power MOSFETS, what's the best way forward? Add .22Ω source resistors?
Now, on a prior repair of a different SA-12, the MOSFETS were not closely matched, and the bias could not be reduced below 460mA. Once I installed a closely-matched quad, I had no problem reducing the bias to 280mA. So, considering the difficulty of sourcing matched power MOSFETS, what's the best way forward? Add .22Ω source resistors?
Trivial solution is to match them of course 😎
Feel free to change R1 to any value needed to pass whatever test current you wish.
That said, I match dozens of MosFets every Month for my Bass and PA power amps (that's why I designed and built this in the first place) and found that those closely matched at 11 mA track quite well at higher currents.
Of course coming from the same batch certainly helps 😉
Modern semiconductor production from one of the big players (Fairchild, ST, IR) is incredibly consistent
https://www.diyaudio.com/community/threads/how-to-match-mosfets-inexpensively.408242/post-7577442
Feel free to change R1 to any value needed to pass whatever test current you wish.
That said, I match dozens of MosFets every Month for my Bass and PA power amps (that's why I designed and built this in the first place) and found that those closely matched at 11 mA track quite well at higher currents.
Of course coming from the same batch certainly helps 😉
Modern semiconductor production from one of the big players (Fairchild, ST, IR) is incredibly consistent
https://www.diyaudio.com/community/threads/how-to-match-mosfets-inexpensively.408242/post-7577442
Last edited:
Reduce R13 (~100k?) or R60 (~5.1k?) and you should be able to reduce the idle current to zero. I'd add a shunt resistor to one of these that is about 5x (~470K||100K=82K).
Note that this bias circuit is a bit dangerous because a corroded bias pot may go open and over bias the FETs into destruction. A quick fix is to connect the pot slider to the top pin.
The 0.22 Ohm source resistors may be a fix to reliability issues since it will improve the thermal stability. I suspect the maker added these for that reason.
Note that this bias circuit is a bit dangerous because a corroded bias pot may go open and over bias the FETs into destruction. A quick fix is to connect the pot slider to the top pin.
The 0.22 Ohm source resistors may be a fix to reliability issues since it will improve the thermal stability. I suspect the maker added these for that reason.