P O W E R ! ! ! 
What transformer power would you use with an amplifier of 2x75W? I see that you like really powerfull power supplies,so i am curious what is your opinion about psu for an amplifier with that output power.
What Vac do you mean-on the output to a speaker or... ?
Many people have different wiews about power supply for AB-class amplifier and i like to hear them all. My opinion would be: transformer not under 1,5 times total output power (2 x is totaly all right) and not under 2000uF per 1A through speaker /per rail. Not that i have done any calculations or listening tests but just by reading on forums opinions of the people that were willling for writing. Some say that even 500uF per 1A through speaker is all right...
Many people have different wiews about power supply for AB-class amplifier and i like to hear them all. My opinion would be: transformer not under 1,5 times total output power (2 x is totaly all right) and not under 2000uF per 1A through speaker /per rail. Not that i have done any calculations or listening tests but just by reading on forums opinions of the people that were willling for writing. Some say that even 500uF per 1A through speaker is all right...
If you are referring to post83, then read again and you will discover that the post is dealing with transformer specification.
With one bridge rectifier:
Vdc/1,41 + 0,7= Vac
With two bridge rectifiers:
Vdc/1,41 + 1,4= Vac
Voltage is not a problem,that is something that all agree about. Power of transformer and needed capacitance in PSU is something that most people do not agree about.
Thanks anyway.
Power supply is transformer + bridge rectifier + smoothing capacitance.
The transformer for good economic performance needs to have a VA approximately from one times to two times the total maximum output power.
The bridge rectifier needs to have voltage and current rating to be reliable over many years of repeated cold starts and continuous operation.
The smoothing capacitance needs to have a voltage rating that exceeds the worst case voltage that could occur on the PSU and sufficient capacitance to give a ripple performance that matches the requirements of the amplifier.
No specific numbers are given because you have not given any information about what you want.
The transformer for good economic performance needs to have a VA approximately from one times to two times the total maximum output power.
The bridge rectifier needs to have voltage and current rating to be reliable over many years of repeated cold starts and continuous operation.
The smoothing capacitance needs to have a voltage rating that exceeds the worst case voltage that could occur on the PSU and sufficient capacitance to give a ripple performance that matches the requirements of the amplifier.
No specific numbers are given because you have not given any information about what you want.
I did build SR200 earlier according to instructions provided by Miles and it worked very fine. But that was about 3 to 4 year ago. And what is sufficient information. Kindly let me know what I am missing so I can learn and ask more intelligent question in future if necessary.
that is because, your PSU will have some power left when you switch off your amp and it gets used up. as Andrew indicated, to avoid this you would need Speaker Protection Circuit with relay built in - so that speaker out is disconnected the moment you switch the amp off.
in general a simple amp includes [some of the components might be optional and some missing like DC Offset Servo!]:
soft start module -> transformer -> rectifier -> DC filter circuit and power reservoir caps -> amplifier pcb -> speaker protection module -> speaker output
it'll include wiring and grounding methods as well.
since you've built an amp, probably you already know this much. if not , please read up on the above chain. even i started from scratch and recently completed my build after reading up on the above topics in this very forum and Rod Elliott's site. and ofcourse with lots of inputs and suggestions from forum members.
actual parameters wrt to input voltage/power ratings, bias current etc are specific each schema/build. that you would have to read up on the forum and refer to the build manual if there is one.
all the best!
Last edited:
Hi all,
I have been reading about the amplifiers of mr. Bora and I'm quite interested in making the Studio SE. I've read some posts about some mistakes found on the pcb files and I got confused. Are the files provided by mr. Bora's site correct or if not, does someone have the correct ones? I would also appreciate any help given for it's construction. I am currently using a Dartzeel NHB108 clone which is quite good, but mr. Bora's designs got my attention. Thanks in advance!
I have been reading about the amplifiers of mr. Bora and I'm quite interested in making the Studio SE. I've read some posts about some mistakes found on the pcb files and I got confused. Are the files provided by mr. Bora's site correct or if not, does someone have the correct ones? I would also appreciate any help given for it's construction. I am currently using a Dartzeel NHB108 clone which is quite good, but mr. Bora's designs got my attention. Thanks in advance!
The Universal-150 is almost the same as the sample the LTC spice ...examples\Educational\100W.asc.
However I found it could be improved in a couple ways.
1. Most important is that it needs a boosted voltage for the drivers because the FETS can not being driven close to the rails. My favorite solution is a ~30% bootstrap on the driver rails. That way you get 140 Watts @8 from +/-50VDC, or 100W at +/-42VDC
2. The bipolar VAS only makes the bias complicated and adds nothing in the way of improved distortion. There is a variation on that circuit that uses a double current mirror instead of the two 4.7K resistors so that the VAS current is independent, but it does not improve performance.
3. Driving the FETS directly requires a lot of current in the VAS, ie hot MJE340/MJE350, or a serious slew rate limitation.
4. The LTC circuit only uses one pair of FETs but I think you need two pair like the Universal-150, even at 2/3 the voltage.
5. A Darlington VAS improves THD by 10x, but may make it less stable.
However I found it could be improved in a couple ways.
1. Most important is that it needs a boosted voltage for the drivers because the FETS can not being driven close to the rails. My favorite solution is a ~30% bootstrap on the driver rails. That way you get 140 Watts @8 from +/-50VDC, or 100W at +/-42VDC
2. The bipolar VAS only makes the bias complicated and adds nothing in the way of improved distortion. There is a variation on that circuit that uses a double current mirror instead of the two 4.7K resistors so that the VAS current is independent, but it does not improve performance.
3. Driving the FETS directly requires a lot of current in the VAS, ie hot MJE340/MJE350, or a serious slew rate limitation.
4. The LTC circuit only uses one pair of FETs but I think you need two pair like the Universal-150, even at 2/3 the voltage.
5. A Darlington VAS improves THD by 10x, but may make it less stable.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.