I also have some more left. You can read the gb thread here:
http://www.diyaudio.com/forums/showthread.php?t=134554&page=91
Power Supply
Hey Andrew T,
So what your saying is -- if the power supply consisted of a 600VA transformer followed by a 20uF cap followed by R (or L) followed by the big 68uF cap (CRC or CLC on each rail of course), then a thermistor on the primary side of the transformer would be sufficient and that a thermistor on the secondary side would not be necessary.
Is this correct?
Thanks,
Steve
Hey Andrew T,
So what your saying is -- if the power supply consisted of a 600VA transformer followed by a 20uF cap followed by R (or L) followed by the big 68uF cap (CRC or CLC on each rail of course), then a thermistor on the primary side of the transformer would be sufficient and that a thermistor on the secondary side would not be necessary.
Is this correct?
Thanks,
Steve
Nelson Pass has in his F4, F5 power supply 120,000uf. He puts 2 thermistors on the primary side. None on secondary. Having 88,000uf on each rail rather than just 60,000uf on each rail really wont change that. The caps will suck current pretty hard but the thermistor on the primary side is limiting current so there should be no need for anything extra.
Uriah
Uriah
Power Supply
Hey Uriah,
I've built a power supply exactly to Nelson Pass's specs (as recommended for the F5) and it works great. But as per Andrew T's post 3008 on P301, I would think that adding the second thermistor (or resistor) on the secondary side would make life much easier on the rectifier and caps at start-up.
The primary side thermistor is out of the way very quickly (less than half a second). But it seems that the surge on the rectifier and caps could go on for few seconds. So as per the post, I don't think that the primary side thermistor is in the circuit long enough to thoroughly soft-strt the rectifier and caps. By adding the secondatry side thermistor, it would allow the caps to charge over a longer period - which would be easier on them.
I am assuming that up to a certain capacitance, it's probably not worth the effort of adding the second thermistor. But when you use a lot of capacitance, it probably becomes more of an issue. So, I was just wondering at what amount of capacitance does adding the second thermistor become somewhat necessary? And, what cold resistance value of the thermistor is typically used in that location?
Thanks,
Steve
Hey Uriah,
I've built a power supply exactly to Nelson Pass's specs (as recommended for the F5) and it works great. But as per Andrew T's post 3008 on P301, I would think that adding the second thermistor (or resistor) on the secondary side would make life much easier on the rectifier and caps at start-up.
The primary side thermistor is out of the way very quickly (less than half a second). But it seems that the surge on the rectifier and caps could go on for few seconds. So as per the post, I don't think that the primary side thermistor is in the circuit long enough to thoroughly soft-strt the rectifier and caps. By adding the secondatry side thermistor, it would allow the caps to charge over a longer period - which would be easier on them.
I am assuming that up to a certain capacitance, it's probably not worth the effort of adding the second thermistor. But when you use a lot of capacitance, it probably becomes more of an issue. So, I was just wondering at what amount of capacitance does adding the second thermistor become somewhat necessary? And, what cold resistance value of the thermistor is typically used in that location?
Thanks,
Steve
By mF, Andrew was meaning milliFarad, which equals 1000uF (microFarad).
i.e. 20,000uF and 68,000uF.
I was under the impression that a cap tries to charge in one cycle. If thats the case then a current limiter that takes half a second to let the cap 'soft charge' gives it a very long time to charge.
I have Mr.Pass F5 power supply in a chip amp with the thermistors in place the same way and using a 400VA 18V toroid. For some reason I get absolutely zero turn on thump and absolutely zero turn on click even. The speakers dont move a bit and DC offset is at 1.2mV. How lucky can a guy get!? But I think it speaks to the ability of those CL60's in the right place being able to clamp very nicely on the current. By the way I only have 60,000uF, not 120,000uF.
Uriah
I have Mr.Pass F5 power supply in a chip amp with the thermistors in place the same way and using a 400VA 18V toroid. For some reason I get absolutely zero turn on thump and absolutely zero turn on click even. The speakers dont move a bit and DC offset is at 1.2mV. How lucky can a guy get!? But I think it speaks to the ability of those CL60's in the right place being able to clamp very nicely on the current. By the way I only have 60,000uF, not 120,000uF.
Uriah
Power Supply
Hi metalman,
Forgive my typo - I meant to write 20000uF and 68000uF. That was my error.
My question however was twofold:
1. At what capacitance should we be looking to add thermistors on the secondaries.
and
2. What cold resistance value thermistors would we put on the secondary side? Would we use the same 10 ohm thermistors that we are using on the primaries? Or, another value?
Thanks again,
Steve
Hi metalman,
Forgive my typo - I meant to write 20000uF and 68000uF. That was my error.
My question however was twofold:
1. At what capacitance should we be looking to add thermistors on the secondaries.
and
2. What cold resistance value thermistors would we put on the secondary side? Would we use the same 10 ohm thermistors that we are using on the primaries? Or, another value?
Thanks again,
Steve
Hi,
from the mains:- it's peak drive voltage divided by acceptable peak current.
From the secondary:- again it's peak drive voltage divided by acceptable peak current.
It is very unlikely when one enters those numbers into the two equations that one will arrive at the same resistance for both sides of the transformer.
Isn't it so much quicker to type 20mF rather than 20000uF or 12pF rather than 0.0000000012uF? oops should that be 0.000012uF?
One does not have to go back and check that the correct number of zeros have been entered.Same when we come to read the posted value.
I hate having to count all the leading zeros and following zeros when the metric system has made it so easy to avoid those kind of typing and reading errors.
from the mains:- it's peak drive voltage divided by acceptable peak current.
From the secondary:- again it's peak drive voltage divided by acceptable peak current.
It is very unlikely when one enters those numbers into the two equations that one will arrive at the same resistance for both sides of the transformer.
Isn't it so much quicker to type 20mF rather than 20000uF or 12pF rather than 0.0000000012uF? oops should that be 0.000012uF?
One does not have to go back and check that the correct number of zeros have been entered.Same when we come to read the posted value.
I hate having to count all the leading zeros and following zeros when the metric system has made it so easy to avoid those kind of typing and reading errors.
Last edited:
Maybe you guys could study the powersupply schematics of some of Papa's earlier work.
Example :
I'm working on an Aleph1.0 version with a cascoded JFET input stage and an output stage with 50 Hitachi TO3s per channel.
(since none of you guys seemed to want some of them i volunteered for a few hundred of the 2SK135, all-same batch number in a box => http://www.diyaudio.com/forums/showpost.php?p=1566603&postcount=25
)
The Push-Pull capable Aleph1 draws 500W/ch idle from 67V rails, the powersupply of the amp also merely has 2 thermistors on the primaries.
The capacitors in the Aleph1 size a total of 100KuF per channel.
These hold nearly 8 times as much energy at 67V than your F5 capacitors at 24V.
Example :
I'm working on an Aleph1.0 version with a cascoded JFET input stage and an output stage with 50 Hitachi TO3s per channel.
(since none of you guys seemed to want some of them i volunteered for a few hundred of the 2SK135, all-same batch number in a box => http://www.diyaudio.com/forums/showpost.php?p=1566603&postcount=25
)The Push-Pull capable Aleph1 draws 500W/ch idle from 67V rails, the powersupply of the amp also merely has 2 thermistors on the primaries.
The capacitors in the Aleph1 size a total of 100KuF per channel.
These hold nearly 8 times as much energy at 67V than your F5 capacitors at 24V.
