I was skeptical of measuring resistance of a coil of wire.
Coulda shoulda brought the scope home (and a power supply), but it’s too late now. I can’t go and get it either now.
No amp meter on the power supply, but it immediately shuts off if 14 amp is exceeded. All I have right now is a basic multimeter.
Coulda shoulda brought the scope home (and a power supply), but it’s too late now. I can’t go and get it either now.
No amp meter on the power supply, but it immediately shuts off if 14 amp is exceeded. All I have right now is a basic multimeter.
If you have a headlamp that has both filaments intact, run them in parallel to provide more current.
Insert it in the B+ line. If the amp powers up through it, you'll see that a small amount of audio will make it light more.
With no audio, twist the inductors slightly. If one is intermittently shorted, the brightness of the light will vary significantly when twisted.
Insert it in the B+ line. If the amp powers up through it, you'll see that a small amount of audio will make it light more.
With no audio, twist the inductors slightly. If one is intermittently shorted, the brightness of the light will vary significantly when twisted.
Big thanks to Perry and AlbertLeRoux again! I got my amp back to working. 50 bench top power ups and no DC offsets and no AC voltages (w/o signal input). Plays music fine again. Ultimately, the failure was probably my fault due to lack of cleanliness during reassembly - which I'll leave at that. Make sure to take your time with these repairs, clean the board, and carefully inspect everything for debris before reassembling.
I installed another set of output FETs, this time while fastened to the heatsink (this amp is all separate metal components which builds into the fireproof box) and while keeping them cool to the touch during installation.
I also replaced 2 new Panasonic ECWF2W105JA film caps as AlbertLeRoux recommended. Found out the hard way these are very sensitive to too much heat during install - the C4004 immediately started to bulge out and overheat when I turned the amp on. It did not self-heal. Turns out polypropylene caps are typically happy when their top and bottom temperatures do not exceed 150 C (which is for the production line to thermocouple-up and test out with the through-hole process), so quick solder work with no dwelling is required to keep the temp down. This thing almost went in the garbage, but worked fine after putting another cap in C4004 with a quick solder at lower temperature. Get it right on this board to avoid re-work - the ECWF2W105JA leads are thicker and mostly fill the throughholes. Its painful to desolder those legs..
Driver ICs also were carefully removed, tested for resistance against an unused IRS20956S, and confirmed good before a very careful reinstall. I wasn't happy with how they looked before and wanted to rule them out, so now the solder looks professional after using a needle-tip on each pin with good wicking. Below are ballpark resistances for a fresh, brand new IRS20956S:
COM to LO = ~12 M ohm
LO to VCC = ~19 M ohm
COM to VCC = 139 k ohm
VB to HC = ~17.5 M ohm
HO to VS = ~22 M ohm
VB to VS = ~1.8 M ohm
Thanks again!
I installed another set of output FETs, this time while fastened to the heatsink (this amp is all separate metal components which builds into the fireproof box) and while keeping them cool to the touch during installation.
I also replaced 2 new Panasonic ECWF2W105JA film caps as AlbertLeRoux recommended. Found out the hard way these are very sensitive to too much heat during install - the C4004 immediately started to bulge out and overheat when I turned the amp on. It did not self-heal. Turns out polypropylene caps are typically happy when their top and bottom temperatures do not exceed 150 C (which is for the production line to thermocouple-up and test out with the through-hole process), so quick solder work with no dwelling is required to keep the temp down. This thing almost went in the garbage, but worked fine after putting another cap in C4004 with a quick solder at lower temperature. Get it right on this board to avoid re-work - the ECWF2W105JA leads are thicker and mostly fill the throughholes. Its painful to desolder those legs..
Driver ICs also were carefully removed, tested for resistance against an unused IRS20956S, and confirmed good before a very careful reinstall. I wasn't happy with how they looked before and wanted to rule them out, so now the solder looks professional after using a needle-tip on each pin with good wicking. Below are ballpark resistances for a fresh, brand new IRS20956S:
COM to LO = ~12 M ohm
LO to VCC = ~19 M ohm
COM to VCC = 139 k ohm
VB to HC = ~17.5 M ohm
HO to VS = ~22 M ohm
VB to VS = ~1.8 M ohm
Thanks again!
Last edited:
Hi Mbrogz3000, glad you have it up and running again and thanks for the information.
When you get access to an oscilloscope again, compare the rail to rail switching frequency of each of the two speaker terminals. You can probe each terminal, referenced to ground and look at the residual carrier with no audio input.
If the switching frequency is much higher in one compared to the other, it may cause additional stress on the capacitor.
When you get access to an oscilloscope again, compare the rail to rail switching frequency of each of the two speaker terminals. You can probe each terminal, referenced to ground and look at the residual carrier with no audio input.
If the switching frequency is much higher in one compared to the other, it may cause additional stress on the capacitor.
Albert, Mbrogz, you both have commented about the driver IC used in the PDX-F6. Do you happen to have the schematics and could contact me? I have tried to find or even buy the service manual to no avail.
I am also surprised that the Abletech IC is interchangeable with the IRS20956S. The only world where I can imagine this is if Abletechn designed the IC and used IR for a foundry and later gave a licence to IR.
The typical application circuit in the IRS20956S data sheet does not include post filter feedback. Note that the PDX-F6 (which I assume is just the 150 W version of the 100 W PDX-F4) had harmonics < -102 dB and no degradation of THD+N towards higher frequencies, which is better than typical Abletech performance and more like ncore territory. Any signs that those Alpine amps use post filter feedback?
I am also surprised that the Abletech IC is interchangeable with the IRS20956S. The only world where I can imagine this is if Abletechn designed the IC and used IR for a foundry and later gave a licence to IR.
The typical application circuit in the IRS20956S data sheet does not include post filter feedback. Note that the PDX-F6 (which I assume is just the 150 W version of the 100 W PDX-F4) had harmonics < -102 dB and no degradation of THD+N towards higher frequencies, which is better than typical Abletech performance and more like ncore territory. Any signs that those Alpine amps use post filter feedback?
Last edited:
Sorry, this is the link with the PDX-F6 measurements
https://www.audiosciencereview.com/...-pdx-f6-review-4-channel-car-amplifier.28870/
Clearly much better harmonic distortion and also noise performance than published measurements of IRAUDAMPs. It is also interesting that the frequency response is largely independent of load impedance and THD+N vs. power is largely independent of frequency. --> pretty aggressive post filter feedback used?
And all this is in spite of the op amp grave yard preamp upstream that is typcial of car audio amps.
By the way, the PDX-V9 did not do nearly as well:
https://www.audiosciencereview.com/forum/index.php?threads/alpine-pdx-v9-review-car-amplifier.28893/
Noise is good but harmonics are not.
https://www.audiosciencereview.com/...-pdx-f6-review-4-channel-car-amplifier.28870/
Clearly much better harmonic distortion and also noise performance than published measurements of IRAUDAMPs. It is also interesting that the frequency response is largely independent of load impedance and THD+N vs. power is largely independent of frequency. --> pretty aggressive post filter feedback used?
And all this is in spite of the op amp grave yard preamp upstream that is typcial of car audio amps.
By the way, the PDX-V9 did not do nearly as well:
https://www.audiosciencereview.com/forum/index.php?threads/alpine-pdx-v9-review-car-amplifier.28893/
Noise is good but harmonics are not.
Hi guys! can anyone tell me if this chip irs20956s is 100% compatible with this one IRS20957S? Because i don`t find any dealer in Europe for this chip irs20956s or you can tell me another equivalent for this ic. I have the same problem with Aline PDX M6 with 43,2 V at the output speaker @ 12,40V power supply ( at 14,40V power supply is 50,4V at speaker output) PS . I don`t find any data sheet for irs20956s that`s i don`t know what ic is equivalent with irs20956s . Any help is appreciated!