Latest project 300WRMS vertical mosfet amp with speaker protection circuit.
Basic dual LTP front end.
Second LTP has current mirror and Vbe multiplier into 4 vertical mosfets.
Speaker protection is done by a PIC micro controlling a relay while monitoring amp output volts.
If PIC sees DC for 500Ms it turns off relay.
There is also a 3 second delay on relay switching on to stop power up thumps.
Basic dual LTP front end.
Second LTP has current mirror and Vbe multiplier into 4 vertical mosfets.
Speaker protection is done by a PIC micro controlling a relay while monitoring amp output volts.
If PIC sees DC for 500Ms it turns off relay.
There is also a 3 second delay on relay switching on to stop power up thumps.
An externally hosted image should be here but it was not working when we last tested it.
Hi Nigel,
Looks cool, however ... did you try to run it at 300W RMS continuously?
I mean - that would be cool if those 4 guys on a heatsink can handle it, but ... 300W is A LOT of heat dissipation. I would say, it sounds like a bit too much for 2 pairs of IRFP240/9240 devices (correct me if I'm wrong).
Cjeers,
Valery
It does need quite a big heat sink for full power continuously.
If I remember correctly you assume that 2/3 of power supply voltage is average.
This gives 33 volts times 8 amps which is about 250 watts or 125 watts per mosfet.
However mosfets are only on for 1/2 of time as music is AC.
This gives 63 watts dissipation in each mosfet.
Correct me if I missed something or got it wrong.
Bias current is around 10ma per mosfet so doesnt add much to heat.
I tested it today and it gives -3db at 50KHz so it is adequate frequency response.
If I remember correctly you assume that 2/3 of power supply voltage is average.
This gives 33 volts times 8 amps which is about 250 watts or 125 watts per mosfet.
However mosfets are only on for 1/2 of time as music is AC.
This gives 63 watts dissipation in each mosfet.
Correct me if I missed something or got it wrong.
Bias current is around 10ma per mosfet so doesnt add much to heat.
I tested it today and it gives -3db at 50KHz so it is adequate frequency response.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Yes its based on the Hitachi datasheet circuit for their lateral mosfets.
Maplin used the Hitachi datasheet for one of their mosfet amp designs.
I added decoupling on front end as old circuit had a little hum with no music through the amp.
I added a Vbe multiplier. I changed laterals to verticals and added source resistors.
Maplin used the Hitachi datasheet for one of their mosfet amp designs.
I added decoupling on front end as old circuit had a little hum with no music through the amp.
I added a Vbe multiplier. I changed laterals to verticals and added source resistors.
It's the original Hitachi Latfet app. note topology from 1979 - adapted to Hexfets. Nigel has posted this before: http://www.diyaudio.com/forums/elektor/269125-amplifier-circuit-elektor.html
Here is the speaker protection part of the circuit.
An externally hosted image should be here but it was not working when we last tested it.
Here is the amp with heatsink.
I am also using a PC fan to help with cooling.
I am also using a PC fan to help with cooling.
An externally hosted image should be here but it was not working when we last tested it.
Is that your PCB design?
Nice
Assuming 70% efficiency in a Class AB (little idle dissipation) you have about 30% dissipation, so 300*0.3=90W total, and that at about 40% sinewave power on the rated load; at full power you actually dissipate less and even less if clipping.
FWIW I have used 4 x IRFP250/240 for ages (and earlier used 4 x 2N3773) in my 300W@4 ohms Bass amplifiers with no problems at all.
Hetasink is not a finned one but the amplifier back panel, fan cooled.
Nice
Assuming 70% efficiency in a Class AB (little idle dissipation) you have about 30% dissipation, so 300*0.3=90W total, and that at about 40% sinewave power on the rated load; at full power you actually dissipate less and even less if clipping.
FWIW I have used 4 x IRFP250/240 for ages (and earlier used 4 x 2N3773) in my 300W@4 ohms Bass amplifiers with no problems at all.
Hetasink is not a finned one but the amplifier back panel, fan cooled.
Hi Guys
Nigel, your schematics are incredibly difficult to read. It would be easier if they had a white background. Negative space sucks the light out of the room...
Four devices is fine if the 300W is transient. You will be deaf by the weekend if you have it cranked all the time. Not a great prospect if you like building things you can play music through.
Have fun
Nigel, your schematics are incredibly difficult to read. It would be easier if they had a white background. Negative space sucks the light out of the room...
Four devices is fine if the 300W is transient. You will be deaf by the weekend if you have it cranked all the time. Not a great prospect if you like building things you can play music through.
Have fun
Is that your PCB design?
Nice
.
I have the pcb's made in China.
They are about a third of the price they are in the UK.
Its usually 2 weeks from order to arrival.
The schematics are reduced before putting them online as I have limited web space.
They look fine in the CAD software.
I prefer black background as white gives me migraines and eye strain.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.