Hi
I am rebuilding 2 A370's and 2 P270's and would appreciate help with replacing the output devices.
In rebuilding I have decided to change the output devices as these run very hot. (In the rebuilt ones I will run them at about 50 degree centigrade as the mark 2 versions of each ampifier did.)
At present the amps all contain different output devices as follows;
P270 - 2sk1058 and 2sj 162 ( sound very good )
P270 - 2sj83 and 2sk227
A370 - buz - 905p and buz 901p
A370 - 2sj118 and 2sk413 ( sound good )
I would like to standardise on the output devices so that I can match sets for each amplifier economically and without having to buy different devices for each amplifier. I have a LOT of irfp240 and irfp9240 ( which I like the sound of ) and would prefer to use these if possible - although they have a greater capacitance. Has anyone tried these in the MF amps?
Otherwise any advise on the best FETS to use would be appreciated.
Don
I am rebuilding 2 A370's and 2 P270's and would appreciate help with replacing the output devices.
In rebuilding I have decided to change the output devices as these run very hot. (In the rebuilt ones I will run them at about 50 degree centigrade as the mark 2 versions of each ampifier did.)
At present the amps all contain different output devices as follows;
P270 - 2sk1058 and 2sj 162 ( sound very good )
P270 - 2sj83 and 2sk227
A370 - buz - 905p and buz 901p
A370 - 2sj118 and 2sk413 ( sound good )
I would like to standardise on the output devices so that I can match sets for each amplifier economically and without having to buy different devices for each amplifier. I have a LOT of irfp240 and irfp9240 ( which I like the sound of ) and would prefer to use these if possible - although they have a greater capacitance. Has anyone tried these in the MF amps?
Otherwise any advise on the best FETS to use would be appreciated.
Don
Forget the vertical FETs unless you are tooled up to properly debug the resulting abomination.
1058/162 are easy to get and can run fairly hot, but they are plastic devices so be sensible.
I got 16pairs a while back and manged to get 7 well matched sets of four, the Nchannel are consistently closer than the Pchannel . The 8th set had one device that was a few more mV different (well actually less mVgs) from it's partners and it shows up as extra quiescent current in that odd device.
I can only run 75mA per device otherwise the sinks get too warm.
1058/162 are easy to get and can run fairly hot, but they are plastic devices so be sensible.
I got 16pairs a while back and manged to get 7 well matched sets of four, the Nchannel are consistently closer than the Pchannel . The 8th set had one device that was a few more mV different (well actually less mVgs) from it's partners and it shows up as extra quiescent current in that odd device.
I can only run 75mA per device otherwise the sinks get too warm.
Andrew
Thank you for your reply. It is very helpful.
Have you tried the buz 905 and buz901? If you have any comments on those devices that would aso be helpful.
Don
Thank you for your reply. It is very helpful.
Have you tried the buz 905 and buz901? If you have any comments on those devices that would aso be helpful.
Don
Hafler used Exicon after Hitachi quit making their parts.
http://www.profusionplc.com/subsites/exicon.htm
The plastic only run GBP 4.15~5 in small quantities.
http://www.profusionplc.com/subsites/exicon.htm
The plastic only run GBP 4.15~5 in small quantities.
Hi,AMV8 said:
no experience of any of the other Lateral FETs.
I have heard that they are all very similar and probably made under licence from Hitachi.
The specs from the old To3 devices to the whole range of current devices are VERY close.
I wonder if the same silicon shop is selling wafers off to be cut up and soldered to the backplate of these different "manufacturers"?
Andrew
Thanks again.
I will try the ones you suggested earlier. I might also have a go with the vertical fets since I have a lot already.
In general it is the front end of the P280 and A370 that has an unusual IC arrangement and ( I find ) sensative to changes. The output architecture is fairly standard.
I will be making changes in any case to run the amplifiers at lower temperatures - about 55 degree C - as the MK2 versions did.
Don
Thanks again.
I will try the ones you suggested earlier. I might also have a go with the vertical fets since I have a lot already.
In general it is the front end of the P280 and A370 that has an unusual IC arrangement and ( I find ) sensative to changes. The output architecture is fairly standard.
I will be making changes in any case to run the amplifiers at lower temperatures - about 55 degree C - as the MK2 versions did.
Don
Hi,
Tc=60deg C is very different from running the heatsinks at 60degC.
If the sinks were at 55degC the worst Tc at quiescent could be >=65degC
What I am saying is that 55degC is very hot if you want good SOA from these plastic devices.
Tc=60deg C is very different from running the heatsinks at 60degC.
If the sinks were at 55degC the worst Tc at quiescent could be >=65degC
What I am saying is that 55degC is very hot if you want good SOA from these plastic devices.
Andrew
I hope you do not mind my asking for more informastion but what heatsink temperature would you recommend?
The original versions of the p270 and A370 ran at high heatsink temperatures - around 70 degrees. The mk2 versions ran at around 55 degree heatsink temperatures. I had planned on 55 degrees after rebuilding but I can go lower if that is advisable.
Do you in practice recommend 75ma per device as you mentioned earlier. This would drop the temperature considerably but limits the class A operation.
Out of interest this is about the current that the Dr T Mk2 used and it sounded very good. The circuit on that amplifier was the same as the P270 but with resisters to drop the voltage for the input circuit instead of the pass transistor that the P270 used.
The Dr T Mk2 used 2 output devices instead of 3 in the P270. However the DrT Mk2 sounds a little thin compared to the P270.
Don
I hope you do not mind my asking for more informastion but what heatsink temperature would you recommend?
The original versions of the p270 and A370 ran at high heatsink temperatures - around 70 degrees. The mk2 versions ran at around 55 degree heatsink temperatures. I had planned on 55 degrees after rebuilding but I can go lower if that is advisable.
Do you in practice recommend 75ma per device as you mentioned earlier. This would drop the temperature considerably but limits the class A operation.
Out of interest this is about the current that the Dr T Mk2 used and it sounded very good. The circuit on that amplifier was the same as the P270 but with resisters to drop the voltage for the input circuit instead of the pass transistor that the P270 used.
The Dr T Mk2 used 2 output devices instead of 3 in the P270. However the DrT Mk2 sounds a little thin compared to the P270.
Don
Hi,
how hot your run the FETs is determioned by the loadline (the number of parallel devices) and the temperataure de-rated SOAR. For four pair on +-69Vdc and Tc=50degC the amp can safely drive a 60degee phase angle 8ohm load to about 195W until the heatsink starts to rise even higher.
how hot your run the FETs is determioned by the loadline (the number of parallel devices) and the temperataure de-rated SOAR. For four pair on +-69Vdc and Tc=50degC the amp can safely drive a 60degee phase angle 8ohm load to about 195W until the heatsink starts to rise even higher.
Hi,
had to dash to a meeting.
The 75mA I ran was limited by the dissipation of the sinks.
Most designers say that FETs get better as the bias is raised all the way to full ClassA.
Borbely states 100mA minimum per output pair and 500mA total minimum Ib.
had to dash to a meeting.
The 75mA I ran was limited by the dissipation of the sinks.
Most designers say that FETs get better as the bias is raised all the way to full ClassA.
Borbely states 100mA minimum per output pair and 500mA total minimum Ib.
Andrew
Thanks again for your information. Now I understand your comments on the 75ma.
My experience is largely with class A amp using the IRF fets and also with the Musical Fidelity Dr T Mk2, P270 and the A370 which use every make but IRF fets. Interestingly I have found that the IRF fets do seem better in full class A - I normally use 60 watt full class A amps - but I have found that the MF amplifiers sound good in class A/AB as long as I use about 75ma plus per device. (The speakers I use are direct connected 5" mid range with some assistance at the treble and base and they are a fairly easy load 4 ohm to drive.)
I will have a go with the IRF devices but I will also buy some of the ones you mention and do a comparison. I do have test gear ( old ) so I will take some measurements.
Don
Thanks again for your information. Now I understand your comments on the 75ma.
My experience is largely with class A amp using the IRF fets and also with the Musical Fidelity Dr T Mk2, P270 and the A370 which use every make but IRF fets. Interestingly I have found that the IRF fets do seem better in full class A - I normally use 60 watt full class A amps - but I have found that the MF amplifiers sound good in class A/AB as long as I use about 75ma plus per device. (The speakers I use are direct connected 5" mid range with some assistance at the treble and base and they are a fairly easy load 4 ohm to drive.)
I will have a go with the IRF devices but I will also buy some of the ones you mention and do a comparison. I do have test gear ( old ) so I will take some measurements.
Don
Hi,
I am not sure of the 118/413 pair, but ALL the others used in the MF Ax70series amps are Lateral FETs.
The change to Vertical FETs will completely change the amplifier and particularly the thermal stability of the output stage.
You are likely to blow it up if you change to VFETs.
Have a look at Vgs of a lateral and compare it to a Vertical. They are different animals.
Look at the change in Vgs with temperature. Again they are completely different.
One cannot just swap between LFets and VFets. This becomes a redesign.
I can tell you what Fets NOT to use in an MF amplifier. Don't use IRF!!!!!
I am not sure of the 118/413 pair, but ALL the others used in the MF Ax70series amps are Lateral FETs.
The change to Vertical FETs will completely change the amplifier and particularly the thermal stability of the output stage.
You are likely to blow it up if you change to VFETs.
Have a look at Vgs of a lateral and compare it to a Vertical. They are different animals.
Look at the change in Vgs with temperature. Again they are completely different.
One cannot just swap between LFets and VFets. This becomes a redesign.
I can tell you what Fets NOT to use in an MF amplifier. Don't use IRF!!!!!
fets
Hi you can replace all the fets with 2sk2221 and 2sj352.
thay are pin compatible with old plastic lateral fets.
if the amp has had the semilab fets fitted (exicon) you will need to lower the bius current.
thay are more stable than the 2sk1058 2sj162 and the exicon's
i get these fets in reguly.
these are the plastic equvilent of the metal 2sk176 2sj56 8 amp 200v
andy
Hi you can replace all the fets with 2sk2221 and 2sj352.
thay are pin compatible with old plastic lateral fets.
if the amp has had the semilab fets fitted (exicon) you will need to lower the bius current.
thay are more stable than the 2sk1058 2sj162 and the exicon's
i get these fets in reguly.
these are the plastic equvilent of the metal 2sk176 2sj56 8 amp 200v
andy
output fets
hi there im using buz900 buz905 pairs on my mf a370 (metal packages) got them from ebay (there from semelabb ) 10 pairs of each for £75 . these have been running at 80mv and sound fantastic .. ps do you find the a370 has some lift at the bottom end ??????????
hi there im using buz900 buz905 pairs on my mf a370 (metal packages) got them from ebay (there from semelabb ) 10 pairs of each for £75 . these have been running at 80mv and sound fantastic .. ps do you find the a370 has some lift at the bottom end ??????????
- Status
- Not open for further replies.