IRFP240/9240 Amplifier (simulated on TINA)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Here is my progress:

I'm still waiting for the magnetic wire to arrive.

I need to know a few things:

What is the bias value?

Where do I measure the bias?

Bias rises clockwise or anti-clockise at the trimmer?
 

Attachments

  • image.jpg
    image.jpg
    820.3 KB · Views: 399
Founder of XSA-Labs
Joined 2012
Paid Member
Here is my progress:

I'm still waiting for the magnetic wire to arrive.

I need to know a few things:

What is the bias value?

Where do I measure the bias?

Bias rises clockwise or anti-clockise at the trimmer?

Nice work!

For bias adjustment, see post 270. Put DMM test leads on points 1-1 and set voltage across 0.22R+0.22R for 80mA (about 35mV). Double check across points (2-2) and should be double at about 70mV.

Please use two 10R, 10W safety resistors (one each) between power supply rails and amp and monitor voltage across that during initial turn on to make sure nothing is terribly wrong. It will limit current so you don't fry the outputs in an instant. I think CCW increaases bias if I remember right - not critical as you will see effect as you turn screw.

http://www.diyaudio.com/forums/soli...-amplifier-simulated-tina-27.html#post4726968

551548d1464268009-irfp240-9240-amplifier-simulated-tina-z02-pcb-silk-02-id.jpg
 
Last edited:
Bias should be around 80mA. Measure the voltage drop across a pair of emitter resistors to figure out the bias. If you are using 0.22R emitter resistors, you should read 36mV.

Set the bias pot to it's highest resistance. You have them installed in opposite directions, so setting will be in opposite directions to each other. Set the offset pot to around the center
 
First, double check all your transistor orientation and location. With power disconnected, measure the resistance of each power rail to ground, and each power rail to your heat sinks.

It looks like you may have Q9 & Q10 installed backwards. The square pads are transistor base, which is pin 3 on a TO-126.

Use a bulb limiter for initial testing to stop smoking parts too.
 
Last edited:
Nice work!

For bias adjustment, see post 270. Put DMM test leads on points 1-1 and set voltage across 0.22R+0.22R for 80mA (about 35mV). Double check across points (2-2) and should be double at about 70mV.

Please use two 10R, 10W safety resistors (one each) between power supply rails and amp and monitor voltage across that during initial turn on to make sure nothing is terribly wrong. It will limit current so you don't fry the outputs in an instant. I think CCW increaases bias if I remember right - not critical as you will see effect as you turn screw.

http://www.diyaudio.com/forums/soli...-amplifier-simulated-tina-27.html#post4726968

551548d1464268009-irfp240-9240-amplifier-simulated-tina-z02-pcb-silk-02-id.jpg

Everything's right, but the voltage between 1-1 and 2-2 should be the same (35-36mV) ;)
Ah Jeff, noticed your previous post too late :)
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
One of my amps initially had a very high railed output and finally traced it to a cold solder joint at R20 (22R 1W). When everything is soldered and placed right, this amp fires up with stable bias and offset. Took a day an removing a bunch of transistors to track that down though. So look for suspect SMT solder pads.

@vzaichenko and jwilhelm: sorry I messed up that 2-2 probe point for the bias - so thanks for catching that. I remember there was a point where the voltage was doubled, imagining it?
 
One of my amps initially had a very high railed output and finally traced it to a cold solder joint at R20 (22R 1W). When everything is soldered and placed right, this amp fires up with stable bias and offset. Took a day an removing a bunch of transistors to track that down though. So look for suspect SMT solder pads.

@vzaichenko and jwilhelm: sorry I messed up that 2-2 probe point for the bias - so thanks for catching that. I remember there was a point where the voltage was doubled, imagining it?

Many people recommend measuring bias across a single emitter/source resistor, which would be 17-18mV. Valery's recommendation is across two emitter resistors, which doubles the value.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.