The MJE+MJL looks and measures really nice for the CCS. But they are not available where I live, so I think I will have to wait more than I thought I would. I think I'll try it with whatever alternatives available here. Don't know why but a vast variety of MOSFET from many manufacturers can be found in almost any electronics shop here. A shame that they proved themselves unfit for the CCS in the tests. But it's okay.
shaan -
what power BJTs are available at your place?
I am afraid that the MOSFET CCS would be difficult to manage without an oscilloscope (wideband if possible).
Regards,
what power BJTs are available at your place?
I am afraid that the MOSFET CCS would be difficult to manage without an oscilloscope (wideband if possible).
Regards,
I usually find most of the general purpose BJTs here and many power BJTs too. The thing is that driver transistors with ft like 30MHz are usually costly and shops usually don't stock them unless there is enough demand from buyers.
Possibly they won't bring any from the BIG distributors even if I pay them in advance and place an order. It's a shame to say but they don't seem to care about DIY stuffs at all and one can see blackout on their faces when asked for something which is either not used in the TVs, radios or have been released recently. Much less they care for new and better products in the market. It's Fun. 😀
shaan,
I believe you could use slower drivers than I did. Just give it a try, and you can use a FM receiver as a sensor again.
Best,
I believe you could use slower drivers than I did. Just give it a try, and you can use a FM receiver as a sensor again.
Best,
Where in India you live - I have seen many MJE or MJL parts are available in Bangalore SJP road area. YOU could ask some of the members( I have seen few in this forum) who live in Bangalore area to procure and mail it to you.
since shipping from US may not be economical - try either Bangalore or Pune /Mumbai area for these parts
kannan
I live near Kolkata. Those parts are not available here yet but I think soon they will be. I think the only way to have them asap is to go Bangalore/Mumbai and bring them myself. I am planning a tour in January, so I think I'll get them soon.
Thanks for the info.
Last edited:
MJE and KSE parts are available in Bangalore but I have not seen MJL 15xxx parts as yet. However, these are listed in RS and Farnell.
You can also look for equivalent parts from 2SC /2SD japanese as well as BD or such alternates for MJL part. Some shops even can get it for you from overseas. IF you have paypal account you can even buy them from outside india through ebay.
since shipping costs are in the region of about 12$ from US , unless you buy two dozen of them, it is not economical to ship. However if somebody is travelling from US then you can ask them to get it for you - as many today have friends or relatives in the US
kannan
Hello everyone.
The RF problem with the MOSFET CCS is solved. 😀 😎
Today I connected the MPF again and was thinking about the MOSFET CCS for the last time. At the same time I ran some AC analysis of both BJT CCS and MOSFET CCS at the positive of the 1ohm CCS load. The results of both simulations are shown below.
The main change in these schematics is a 1K gate stopper for the gain MOSFET instead of the default 270ohm. The analysis showed a lower magnitude AC at the +ve of the resistor in the case of the MOSFET CCS MPF. I tried it out of curiosity. It is working and is stable and after about 3 hours of music still no oscillation; all FM channels are alive and noise free 😀 and the turn on click is gone too.
I don't know exactly why it is working without oscillation but I think it's the 1K resistor at the gate of the upper fet. As shown in the schematic the CCS fet has no gate stopper.
I don't know how to generate those square wave response curves in Simetrix so am asking all the kind hearted experts here to help me with it. I will be very grateful if someone tests the MOSFET CCS MPF for square wave and posts the picture.
Thanks for your time.
The RF problem with the MOSFET CCS is solved. 😀 😎
Today I connected the MPF again and was thinking about the MOSFET CCS for the last time. At the same time I ran some AC analysis of both BJT CCS and MOSFET CCS at the positive of the 1ohm CCS load. The results of both simulations are shown below.
TIP142 CCS MPF
TIP142 CCS MPF AC
IRFP150N CCS MPF
IRFP150N CCS MPF AC
TIP142 CCS MPF AC
IRFP150N CCS MPF
IRFP150N CCS MPF AC
The main change in these schematics is a 1K gate stopper for the gain MOSFET instead of the default 270ohm. The analysis showed a lower magnitude AC at the +ve of the resistor in the case of the MOSFET CCS MPF. I tried it out of curiosity. It is working and is stable and after about 3 hours of music still no oscillation; all FM channels are alive and noise free 😀 and the turn on click is gone too.
I don't know exactly why it is working without oscillation but I think it's the 1K resistor at the gate of the upper fet. As shown in the schematic the CCS fet has no gate stopper.
I don't know how to generate those square wave response curves in Simetrix so am asking all the kind hearted experts here to help me with it. I will be very grateful if someone tests the MOSFET CCS MPF for square wave and posts the picture.
Thanks for your time.
connect a wavefor generator at the input using :
place/voltage source/waveform generator..
then set the geneartor values,
place a voltage probe at the output using :
probe/place fixed voltage probe..
then run, it works.!!
I think what you are saying is right. But, with my limited knowledge of audio engineering I cannot do more than what I have done.
I used to place a 270ohm resistor between the preamp output and the upper MOSFET's gate and everything seemed to be ok. This was the suggested config in the ESP pages. I did what was said there and the result was great indeed. In the case of the MOSFET CCS I saw in the simulator that the AC in the CCS load resistor is decreased if I increase the value of that 270ohm resistor to 1000ohm. There I tried it in real world and here it looks like it is working. At the same time, the simulator also shows decreased AC in the CCS load in the MPF with darlington CS; much less than that with the default config suggested in the ESP pages. This made me a bit curious and I could not help it. But as it is stable and not oscillating(the fm receiver is ok with this config), I thought it would be a nice idea to tell other people about it so that they might test it out for themselves and tell their own stories and I can be sure that I am not day-dreaming. 😀
thanks.
Last edited:
Yes it was, in the year 2001. Please believe me, the correct position of the gate stopper (R11) is between the zener cathode and gate of the MOSFET. The additional R10 only helps.
first ,set the generator to the test frequency, let s say 10 khz;;
choose transient in the simulator menu options..
set the simulator step to 10n
click "output at print step"
time of analysis is 1/F mini, in this case 100 uS...
I ran 20 KHz squarewave tests to both circuits and here are the results.
With 20 KHz squarewave the current fluctuation in the load resistor of the BJT CCS is more than 400mA, while in case of the FET CCS it is about 140mA. This looks like an advantage of the FET CCS over the Darlington CCS. Besides, the MOSFET CCS sounds just as good as the old one, equally in bass, mid and treable.
With TIP142 CCS
With IRFP150N CCS
With IRFP150N CCS
With 20 KHz squarewave the current fluctuation in the load resistor of the BJT CCS is more than 400mA, while in case of the FET CCS it is about 140mA. This looks like an advantage of the FET CCS over the Darlington CCS. Besides, the MOSFET CCS sounds just as good as the old one, equally in bass, mid and treable.
- Status
- Not open for further replies.