Good day Group! I am trying to upgrade my MF A-370 clone.
In this instance, I am looking at the VAS and driver stages (see attached schematic diagram), considering higher power BJTs. Not sure this would give significant improvements, would appreciate some help from the Group experts.
In the VAS stage (the BJTs are in common-base configuration) I am considering to replace the MPSA43/93 with KSC3503/KSA1381, capable of higher power. Is this a good idea? Any other device could be better?
The gate driver stage (emitter-follower configuration) uses BC414c/BC416c.
I understand the power mosfet gates would be better driven by higher power devices due to their capacitive nature (I am using 2sk135/2SJ50, 900 and 600 pF). Also here I am considering suitable higher power BJTs (TO126 or TO220). I searched the forum but did not find any clear suggestion. Would therefore appreciate some help.
Thanks and Regards, Giovanni
In this instance, I am looking at the VAS and driver stages (see attached schematic diagram), considering higher power BJTs. Not sure this would give significant improvements, would appreciate some help from the Group experts.
In the VAS stage (the BJTs are in common-base configuration) I am considering to replace the MPSA43/93 with KSC3503/KSA1381, capable of higher power. Is this a good idea? Any other device could be better?
The gate driver stage (emitter-follower configuration) uses BC414c/BC416c.
I understand the power mosfet gates would be better driven by higher power devices due to their capacitive nature (I am using 2sk135/2SJ50, 900 and 600 pF). Also here I am considering suitable higher power BJTs (TO126 or TO220). I searched the forum but did not find any clear suggestion. Would therefore appreciate some help.
Thanks and Regards, Giovanni
Attachments
I think that you'll get potentially better "update" with another opamp. But beware of oscillations, compensation should be tweaked etc.
Thank you Bogdan, I am after all improvement recommendations, so also for the input IC. It seems I could replace the LM318N with ICs such as NE5534AN or OPA627. I understand these require to isolate pin 1, 5 and 8 (all compensation pins).What do you think?
Going back to my initial question, I am considering the replacement of the drivers (bc414c/416c) based on other comments I red about the need of higher power driver for the mosfet gates. Any views on this?
Also, more in general, any other suggestion?
I am not an expert, this is one of my hobbies, so I ask to the many real experts present in this Group.
Thanks, Giovanni
Going back to my initial question, I am considering the replacement of the drivers (bc414c/416c) based on other comments I red about the need of higher power driver for the mosfet gates. Any views on this?
Also, more in general, any other suggestion?
I am not an expert, this is one of my hobbies, so I ask to the many real experts present in this Group.
Thanks, Giovanni
BC414/416 is just a small TO-92 transistor
So using KSC3503 and KSA1381
in this positions would make more sense
for higher current
The Vas transistors MPSA 43/93
are pretty common general purpose transistors and rather slow
around 50 Mhz
Just using 2N5551 / 5401 would improve performance
Since they are around 300 Mhz
Of course as noted the LM318
has a bandwidth of 15 Mhz and a Slew Rate of 50V/us
So no the 5534 with 10 Mhz and Slew Rate of 13V/us
would not outperform it
OPA627 has 16 Mhz and Slew Rate of 55V/us
so would be slight improvement
Likely not notable because the common base driven Vas
limits the high frequency THD performance of the amp in general.
Of course a very high speed Opamp which is not compensated
and require some changes would be a LM7171
200 Mhz Bandwidth and 4100V/us slew rate
So using KSC3503 and KSA1381
in this positions would make more sense
for higher current
The Vas transistors MPSA 43/93
are pretty common general purpose transistors and rather slow
around 50 Mhz
Just using 2N5551 / 5401 would improve performance
Since they are around 300 Mhz
Of course as noted the LM318
has a bandwidth of 15 Mhz and a Slew Rate of 50V/us
So no the 5534 with 10 Mhz and Slew Rate of 13V/us
would not outperform it
OPA627 has 16 Mhz and Slew Rate of 55V/us
so would be slight improvement
Likely not notable because the common base driven Vas
limits the high frequency THD performance of the amp in general.
Of course a very high speed Opamp which is not compensated
and require some changes would be a LM7171
200 Mhz Bandwidth and 4100V/us slew rate
Replacing a 318 input stage is not straightforward - most other amps I’ve seen that use them are better off leaving the 318 in there. Especially if it’s still good - replacements are hard to come by. Anything else usually requires much different compensation to get a stable amp.
The slew rate is limited by the charge/discharge of the mosfet gate capacitance anyway. You would have to alter resistor values in the vas in order to get much more out of it (especially falling edge) in addition to making those transistors beefier. I would have been worried about low power devices overheating or just driven outside SOA when hit with sustained high frequency signal. Those TO126 devices are what I would have chosen in the first place, if you’re going to do that get them while you can - everything in that package that has any kind of fT is on its way out of production. In 10 years you won’t even be able to dream about getting C3503/A1381. Everything as fallen one by one already.
The slew rate is limited by the charge/discharge of the mosfet gate capacitance anyway. You would have to alter resistor values in the vas in order to get much more out of it (especially falling edge) in addition to making those transistors beefier. I would have been worried about low power devices overheating or just driven outside SOA when hit with sustained high frequency signal. Those TO126 devices are what I would have chosen in the first place, if you’re going to do that get them while you can - everything in that package that has any kind of fT is on its way out of production. In 10 years you won’t even be able to dream about getting C3503/A1381. Everything as fallen one by one already.
Thank you wg_ski and WhiteDragon.
I was thinking of using KSC3503 and KSA1381 (TO126) for the VAS, and eventually an even more powerful couple (in TO220) for the mosfet drivers (as per my ad title) for which I have not identified suitable candidates, yet.
Is this reasonable?
If yes I would need suggestions for the mosfet driver couple (TO220).
Eventually I could use KSC3503 and KSA1381 (TO126) also for the mosfet drivers.
Remains the question if this will be a drop-in replacement of the existing couples (MPSA 43/93 as VAS and BC414/416 as mosfer driver stage) or I need to alter resistor values in the VAS as mentioned by wg_ski.
I was thinking of using KSC3503 and KSA1381 (TO126) for the VAS, and eventually an even more powerful couple (in TO220) for the mosfet drivers (as per my ad title) for which I have not identified suitable candidates, yet.
Is this reasonable?
If yes I would need suggestions for the mosfet driver couple (TO220).
Eventually I could use KSC3503 and KSA1381 (TO126) also for the mosfet drivers.
Remains the question if this will be a drop-in replacement of the existing couples (MPSA 43/93 as VAS and BC414/416 as mosfer driver stage) or I need to alter resistor values in the VAS as mentioned by wg_ski.
After some thinking I bought as mosfet gate drivers, the following:
- 2SC5171 (to replace the original BC414c)
- 2SA1930 (to replace the original BC416c)
Hope this is a good choice, a bit of an overkill, though...
Any comment is welcome.
As per the LM318N, I don't know whether to retain it or to replace it, I continue reading different opinions... Any help would be appreciated.
Giovanni
- 2SC5171 (to replace the original BC414c)
- 2SA1930 (to replace the original BC416c)
Hope this is a good choice, a bit of an overkill, though...
Any comment is welcome.
As per the LM318N, I don't know whether to retain it or to replace it, I continue reading different opinions... Any help would be appreciated.
Giovanni
I don't think you understand how all this works, you can't get "bettter" mosfet driver, just sticking biger bjt...
bogdan_borko, you are right, I have no electronic study background and/or long experience. I am trying to learn, so far I have no time for systematic studies, I try to use common-sense and get help to those who know better. This group is full of knowledgeable people for this.
30+ years ago, HiFi was one of my hobbies, I decided to build an MF A-370 clone starting from a schematic diagram and a list of components (a bit crazy, eh?). I could have started from a lesser target, I went big...
It took a lot of time and work and I put together an amplifier which worked and sounded well to me. Maybe I will post this story here, soon or later.
Soon after, I moved abroad for work (I am still abroad), my amplifier is in my basement at home since then. Now, I am close to retiring and would like to resume this hobby. To restart, decided to see if I can improve my old amplifier when I go back home.
I am reading a lot about the MF A-370 and its problems, most due to cheap and undersized components, so my initial plan is to improve the component quality and their ruggedness.
So yes, I plan to use bigger BJTs to start with. Is it wrong? Please advise.
Thank you for your help.
30+ years ago, HiFi was one of my hobbies, I decided to build an MF A-370 clone starting from a schematic diagram and a list of components (a bit crazy, eh?). I could have started from a lesser target, I went big...
It took a lot of time and work and I put together an amplifier which worked and sounded well to me. Maybe I will post this story here, soon or later.
Soon after, I moved abroad for work (I am still abroad), my amplifier is in my basement at home since then. Now, I am close to retiring and would like to resume this hobby. To restart, decided to see if I can improve my old amplifier when I go back home.
I am reading a lot about the MF A-370 and its problems, most due to cheap and undersized components, so my initial plan is to improve the component quality and their ruggedness.
So yes, I plan to use bigger BJTs to start with. Is it wrong? Please advise.
Thank you for your help.
Those followers have to supply the gate current fir the mosfets. True, static gate current is zero, but dynamically it is NOT. The faster you turn them on and off the more current is demanded I= Cvd/dt. The device driving it places some limitation. It may or may not be limited by the device itself, but it’s operating conditions as well. Adjusting the rest of the circuit may be necessary to boost the slew rate. it in any case more current is needed. That takes a bigger device, both to keep the beta high when the current does peak, and to manage any increase in dissipation.
I agree, at least monitor the bias current when doing that, it will rise if it's oscillating, try it first with the light bulb in series with mains, or other kind of protection...
Cheers.