Rather try a DC filter - all transformers will buzz when AC is polluted with a small amount of DC.
Overtightening toroids may loosen the secondary wiring, be careful.
Diodes or rectifier bridge must be on a heatsink. I prefer a proper separate one rather than sharing the main or using the chassis bottom because it will transfer a lot of heat onto the transformer that is normally close by.
One NTC on primary is enough. Do not try to cool it, it needs to be hot to stay at 0 resistance.
Sent from my MediaPad T1 8.0 using Tapatalk
Overtightening toroids may loosen the secondary wiring, be careful.
Diodes or rectifier bridge must be on a heatsink. I prefer a proper separate one rather than sharing the main or using the chassis bottom because it will transfer a lot of heat onto the transformer that is normally close by.
One NTC on primary is enough. Do not try to cool it, it needs to be hot to stay at 0 resistance.
Sent from my MediaPad T1 8.0 using Tapatalk
it never gets to zero resistance. After the high current pulses of startup and charging it cools and the resistance increases back towards the cold value.
It's to avoid this higher resistance that you add a delayed bypass to lower the source impedance seen by the transformer. That increased source impedance is reflected through the transformer to the secondaries.
You might want to read this informative article from ESP:
Earthing (Grounding) Your Hi-Fi - Tricks and Techniques
I use ground loop breakers, i don't know if they are "permissible" in the EU.
SMPS for CLASS-A Amplifiers +-25VDC 26A
Hi Everyone
Any one interested into SMPS for CLASS-A Amplifier? +-25VDC
The SMPS drops 6 Volts per rail at 26A DC output.
Just developed a nice 1000W SMPS specifically designed for CLASS-A with on board Turn ON delay / DC Protection.
Attached a picture of the PCBs under assembly.
Other voltages available also
Thanks!
Hi Everyone
Any one interested into SMPS for CLASS-A Amplifier? +-25VDC
The SMPS drops 6 Volts per rail at 26A DC output.
Just developed a nice 1000W SMPS specifically designed for CLASS-A with on board Turn ON delay / DC Protection.
Attached a picture of the PCBs under assembly.
Other voltages available also
Thanks!
Attachments
Last edited:
Ripple
Hello
Ripple at 2.6A DC is around 200mV (RMS).
With 10000uf extra capacitance per rail, Ripple around 103mV (RMS)
Regards
Hello
Ripple at 2.6A DC is around 200mV (RMS).
With 10000uf extra capacitance per rail, Ripple around 103mV (RMS)
Regards
Hello everyone!
I just have a quick question before i can order my parts for the F5! On the schematics it says the resistors at the input should be 4.75K and 47.5K, but on the homepage where i order my parts they only have 4.7K and 47K... Is it ok to change this value? I tried to search for it in the thread but did not find an answer (although i think a lot of people just use 4.7k?)
Furthermore i wanted to ask which parts are the critical one for audio quality? R1, R13 and R14 are in the signal path, so thats clear. What about the rest?
I read different stuff about the VA value of the trafo. 400VA should be enough but 600VA would be better due to temperature and power, but is a little bit louder (buzzing)? Is this still the common experience?
Thanks for all,
Philip
I just have a quick question before i can order my parts for the F5! On the schematics it says the resistors at the input should be 4.75K and 47.5K, but on the homepage where i order my parts they only have 4.7K and 47K... Is it ok to change this value? I tried to search for it in the thread but did not find an answer (although i think a lot of people just use 4.7k?)
Furthermore i wanted to ask which parts are the critical one for audio quality? R1, R13 and R14 are in the signal path, so thats clear. What about the rest?
I read different stuff about the VA value of the trafo. 400VA should be enough but 600VA would be better due to temperature and power, but is a little bit louder (buzzing)? Is this still the common experience?
Thanks for all,
Philip
Most of us use 1% tolerance E24 values.
That would be 47k ±1% i.e. from 46k53 to 47k47
47k5 is an E96 value. They and E192, are usually specified where 0.1% tolerance is actually required to make the circuit work properly.
Just use the ordinary E24 values for both the 4k7 and the 47k.
http://www.logwell.com/tech/components/resistor_values.html
That would be 47k ±1% i.e. from 46k53 to 47k47
47k5 is an E96 value. They and E192, are usually specified where 0.1% tolerance is actually required to make the circuit work properly.
Just use the ordinary E24 values for both the 4k7 and the 47k.
http://www.logwell.com/tech/components/resistor_values.html
Last edited:
ClassA generally use transformers about 6times to 10times the maximum power output rating of the ClassA amplifier.
The 25W/27W F5 would on that basis require a minimum of 150VA to a maximum of 270VA.
A two channel F5 running from a common transformer would therefore require a minimum of 300VA to a maximum of 540VA.
As a check, compare the DC quiescent current draw to the maximum continuous DC current rating when feeding into a capacitor input filter.
A stereo F5 set to a bias of 1.3Adc would demand a continuous DC current of 2.6Adc from the PSU, I'll ignore the 8mA or so going into the jFETs.
A 300VA 25-0-25Vac transformer has an AC current rating of 300/{25+25} = 6Aac
The maximum continuous DC current from that transformer +PSU when feeding into a capacitor input filter is ~3A. Depending on the manufacturer and their ratings it could be as high as 3.6Adc.
2.6Adc compared to 3Adc means you are drawing 87% of the maximum continuous DC rating.
And compared to 3.6Adc you would be drawing 72% of maximum continuous DC
Both those will make the transformer run pretty warm.
I would use a higher VA rating. I generally recommend for a cool running transformer that you use ~50% of the maximum continuous DC rating, when using a capacitor input filter.
400VA 22-0-22Vac has an AC current rating of 9.09Aac
The maximum continuous DC rating will be around 4.5Adc
2.6Amperes demand is 57% of that 4.5Aac. = cool. and only 48% if the DC rating is 5.4Adc
The 25W/27W F5 would on that basis require a minimum of 150VA to a maximum of 270VA.
A two channel F5 running from a common transformer would therefore require a minimum of 300VA to a maximum of 540VA.
As a check, compare the DC quiescent current draw to the maximum continuous DC current rating when feeding into a capacitor input filter.
A stereo F5 set to a bias of 1.3Adc would demand a continuous DC current of 2.6Adc from the PSU, I'll ignore the 8mA or so going into the jFETs.
A 300VA 25-0-25Vac transformer has an AC current rating of 300/{25+25} = 6Aac
The maximum continuous DC current from that transformer +PSU when feeding into a capacitor input filter is ~3A. Depending on the manufacturer and their ratings it could be as high as 3.6Adc.
2.6Adc compared to 3Adc means you are drawing 87% of the maximum continuous DC rating.
And compared to 3.6Adc you would be drawing 72% of maximum continuous DC
Both those will make the transformer run pretty warm.
I would use a higher VA rating. I generally recommend for a cool running transformer that you use ~50% of the maximum continuous DC rating, when using a capacitor input filter.
400VA 22-0-22Vac has an AC current rating of 9.09Aac
The maximum continuous DC rating will be around 4.5Adc
2.6Amperes demand is 57% of that 4.5Aac. = cool. and only 48% if the DC rating is 5.4Adc
Last edited:
Thank you very much!
I see, i was just worried because the Z-Foil i would like to buy (Charcroft Z-Foil and S-Foil | Hifi Collective) have a tolerance of 0.1% and only E24 values...
I see, i was just worried because the Z-Foil i would like to buy (Charcroft Z-Foil and S-Foil | Hifi Collective) have a tolerance of 0.1% and only E24 values...
