Good morning everyone I have a simple question about choosing the transistor to use in a cap multiplier
If I understand correctly it is good to have a device characterized by a high hfe
How come I see cascades of bjts and practically never darlingtons with hfe=1000 ?
Are there any problems that I don't know about using a single Darlington that provides the same hfe as two or three bjts in series? thanks everyone for the help
If I understand correctly it is good to have a device characterized by a high hfe
How come I see cascades of bjts and practically never darlingtons with hfe=1000 ?
Are there any problems that I don't know about using a single Darlington that provides the same hfe as two or three bjts in series? thanks everyone for the help
Last edited:
A good question... many of the older types we commonly use can have a worse HF performance (a lower fT or transition frequency) than carefully selected discrete parts but it would really come down to choosing the correct parts for the job. For a mains supply with 100/120 Hz ripple it should not be an issue.
Hi thank you sincerely for the very kind and valuable advice I prefer to keep things as simple as possible
One darlington for rail should be sufficient i guess I am focusing the circuit i am attaching
https://sound-au.com/articles/cap-mult-f23.gif
of course using a high hfe darlington instead of the bjts I am more confident now
Just one question In case i will use with LM317 what would be the best option ... before or after the LM317 stage ?
i would like to combine the good performance of the LM317 in the low Hz with the better performance of the cap multi in the higher Hz
One darlington for rail should be sufficient i guess I am focusing the circuit i am attaching
https://sound-au.com/articles/cap-mult-f23.gif
of course using a high hfe darlington instead of the bjts I am more confident now
Just one question In case i will use with LM317 what would be the best option ... before or after the LM317 stage ?
i would like to combine the good performance of the LM317 in the low Hz with the better performance of the cap multi in the higher Hz
Attachments
You need a choke (coil) to get rid of the noise at very high frequencies... so something like C-L-Capmultiplier will work well, especially for the sensitive audio section(s), like DACs and streamers... no need for LM317.
C0 strating value should be around 470-1000uF... but you'll need an osciloscope to get things right (perfect?).
Good luck.
C0 strating value should be around 470-1000uF... but you'll need an osciloscope to get things right (perfect?).
Good luck.
No, he means as he has drawn on the schematic in post #5. I would place the LM317 (for regulation purposes) at the end of the supply, connected to C4.
Basically: Transformer -> Rectifier ->CLC->Cap multiplier->LM317. It’s probably overkill, but if you want regulation, there you go. Do read the datasheet on the LM317 as they have an application example (section 9.2).
Best,
Anand.
Basically: Transformer -> Rectifier ->CLC->Cap multiplier->LM317. It’s probably overkill, but if you want regulation, there you go. Do read the datasheet on the LM317 as they have an application example (section 9.2).
Best,
Anand.
FWIW I would always use a Darlington if one of the common ones fills the duty, why not?
I always have tons of TIP142/147 around because those are the power devices in my Guitar amps, so always have a small drawer or a couple sticks of them within arm´s reach.
For higher current (un likely) or voltage (possible), yes, would build up a discrete one.
I always have tons of TIP142/147 around because those are the power devices in my Guitar amps, so always have a small drawer or a couple sticks of them within arm´s reach.
For higher current (un likely) or voltage (possible), yes, would build up a discrete one.
Alternatively, you could use a common mode choke in both plus & minus connections to the bridge rect. (before the first smoothing cap.
Hi thank you very much indeed This is of great help to me I have a so much better picture now To be honest i am understanding that opamps can be very much all i need And sometimes i read comments like that the bashing of opamp based preamps comes from so so power supply ... like the usual 15V fixed regulators So i am trying to understand how much a cap multiplier or a filtering stage before monolithic regulators could improve the overall performance
I understand now a lot indeed I will follow the sequence you recommend In the final regulation stage i could use also fixed regulators up to +/-24VDC
More than enough for a line level unit I just need 2Vrms at the output with low Z and THD+noise
Hi thanks a lot ! perfect and this is what i needed to know
another good point for using darlingtons. My knowledge is very limited but i wonder how in general darlingtons are not more popular
I mean having the driver in the same chip of the output device allows for a less number of active devices needed
If i understand well they were more popular in commercial british integrated amps And some of them were very well received
Probably there are also drawbacks that i do not know
Hi thanks a lot for the valuable suggestion Unfortunately i cannot find them in the LTSpice default library i am using right now
Actually i do not even know if darlington models are present in the library I think they are mixed with bjts ? i will look for them for sure
They seem very robust units Good !
It starts in the double-digit millihenry range. It may be enough to use a transformer that is "lying around". Just try.
By short-circuiting one of the several windings, the effect can be heard in comparison to pure resistance screening (the inductance collapses and the effective resistance of the coil wire remains). Just always try;-)
By short-circuiting one of the several windings, the effect can be heard in comparison to pure resistance screening (the inductance collapses and the effective resistance of the coil wire remains). Just always try;-)
So many people overlook the fact that the psu is important to the sound. I always think of a power amp as a power supply modulator so the impedance and slew rate of the psu shouldn't be overlooked.