UCD 25 watts to 1200 watts using 2 mosfets

Jump to Latest
#3,941
have fun 😛😀🙂
 

Attachments

  • irs 900d corect artwork print .pdf
    irs 900d corect artwork print .pdf
    351.5 KB · Views: 1,680
  • IRS-900D _ PCB 2 thienchay spaced small artwork.pdf
    IRS-900D _ PCB 2 thienchay spaced small artwork.pdf
    342.2 KB · Views: 1,531
  • irs500 component.jpg
    irs500 component.jpg
    132.5 KB · Views: 3,814
  • ab722e449ca921e4.jpg
    ab722e449ca921e4.jpg
    250.2 KB · Views: 3,727
  • IRS-900D _ Overlay_ checked.pdf
    IRS-900D _ Overlay_ checked.pdf
    104.2 KB · Views: 1,278
  • IRS-900D _ Overlay_ checked.jpg
    IRS-900D _ Overlay_ checked.jpg
    166 KB · Views: 3,649
#3,946
Dinithm said:
An externally hosted image should be here but it was not working when we last tested it.
Click to expand...
I have experienced some weird things with this design. Week ago i put this thing together from junk parts and it worked perfectly. Very clean switching waveforms, i was surprised. Only thing heating up was the random iron powder toroid at the output filter which was obviously bad material for this application. Since this amplifier is driving 4ohm subwoofer i came up with 32µH inductor and 2µF capacitor.

Later on i replaced the overheating toroid with impedance matching choke from my induction heater project which is good up to 125Amps at 40kHz. Big surprise was that when the choke was set to exactly same inductance than the smoking hot one, the amplifier lost its stability.

Problem i noticed is switching frequency and duty cycle are jumping all over the place, and voltage in positive rail is creeping up like 1V/second which means i have DC offset at the speaker output. Now the weird thing is, when i unplug the signal source (mobile phone or laptop, both isolated) the positive rail creeping and instability go away.

No matter what, i could not get the 32µH inductor to work on ferrite core, but it works perfectly on super lossy iron powder. Interestingly though, if i remove the air gap completely, total inductance is around 110µH and the amplifier once again works perfectly, but output impedance is of course nowhere near i want it to be, and the positive rail voltage creepage is still there. In this condition i measured whopping 824W DC input power when i rattled whole building with 15" subwoofer and nothing got hot or exploded 😀
 
#3,949
Hello from Ukraine.
I want to collect the amplifier for the subwoofer.
Stopped on the amplifier IRS-D900. I noticed that there are two types of fees. In post # 3941 are shown both. What is proven and can collect it. What is working?
Detex Audio IRS-900D V 2.0 or Thien Chay pcb???
If you have these cards in the format .lay would be very grateful for your help.
I would be grateful to all who will respond and help thanks in advance.
P.S.
Sorry for my english.
 
#3,954
pergo said:
that is my schematic.
It works very well but you've to tune the circuit with real components.
I prefer the IRS20957 version because there are less components, more protection and best performance.
Click to expand...


Hey pergo,

I ran a LTSpice simulation of the schematics attached which I believe it's yours.
There are 93A current spikes through the final MOSFETS due to simultaneous conduction.

How would you tune the schematics to remove the spikes ?
 

Attachments

#3,958
Hello,

..when I saw this modified schematic done by Joel of DIYsmps, it got me interested to resurrect IRS900D 🙂 and after my initial test of Manoj mod on ApexD200, I made some changes (in red color) from the original design of Detex and also with Joeys' mod. I'd like to use the mid-power TO-220 devices for the output because they are easily available and cheaper too.

Here is the schematic and the pcb design...any thoughts before I construct this amp?
 

Attachments

  • IRS900D.jpg
    IRS900D.jpg
    380.8 KB · Views: 2,620
  • classd new sch.png
    classd new sch.png
    62 KB · Views: 2,477
Last edited:
You must log in or register to reply here.
Top Bottom