Power amp under development

Jump to Latest
#2,881
Whar is yours assessment about this audio amplifier (see image)?
A compact design audio amplifier with: LED indicators show output voltage and headroom; clip limiter; DC, overload and temp protections, balanced inputs, soft start output, audio preamplifiers and variable speed fan with temperature heat sink.

Technical Specifications:
Power in to 4 Ohm load: 400W;
THD = 0,02%;
Power supply: +/-70Vcc

This amplifier is designed by me and costs around 50Euro without VAT.
 

Attachments

  • mx400_01.jpg
    mx400_01.jpg
    66 KB · Views: 1,185
#2,883
Hi,
you could try, but who can guarantee the result?

First question:-
what is the peak transient current that the amplifier will demand from the SMPS when driving a real speaker?

Where does that current come from?

What size of SMPS will be needed to meet that demand?
 
#2,884
That kind of SMPS topology is perfect for delivering high peak currents without problems without having to design for continous high power. It's the same as QSC uses (but with an added PFC) and I believe Crown also have used the same topology.

There is no output inductor to saturate so output current is only limited by transformer, diode and transistor dissipation. If IGBT:s are used high peak currents are no problem at all.

Thanks to the high output peak current capability most energy storage can be moved to the primary side - which is good because for the same energy storage electrolytics are smaller at high voltage.
 
#2,889
zerohead_ak47 said:
its good working
Click to expand...

Good news that you have built a working Nmos mkII. Do you have a PCB layout for this? Anyway how does it sound?

AndrewT said:
add gate resistors to the six FETs.
bypass the 2uH with ~4r0. The 18k in the LTP tail is dropping between 60 & 70 Volts. Try halving this value, maybe 10k.
Click to expand...

Andrew's advice is sound, you should add gate resistors to each FET gate. This will help protect against oscillations.

Assuming 0.6v drop for the diode and transistor junctions the current through the 18k resistor will be 3.33mA equating to a voltage drop of 60v, leaving 10v across the ccs transistor. The resistor could be reduced to 15k leaving 20v across the transistor if desired.

megajocke said:
And source resistors for the FETs in the negative rail
Click to expand...

These resistors should be moved as Megajocke suggests. This will improve the current sharing between the negative rail FETs.

Other thoughts are:

- Put a 0.1uf capacitor across the zener.

- Increase the filter capacitors on FET rails to 1,000uF.

- The 100uF capacitor on the rails feeding the 1st & 2nd stage could cause problems if not grounded correctly. In practice it is not needed so if you have any noise issues try removing it.

Cheers
Quasi
 
You must log in or register to reply here.
Top Bottom