Got a RF P300-2 in to repair, left channel blew, killed the power supply. Got the power supply up and running and repaired the left channel drive, however the fets that it uses are 36P15 (P channel) and 62N15P (N channel). The 62N15P is a monster at 62 amps and is non-stocked at Mouser and Digikey. Normally Rockford pairs the 36P15 with the 28N15. Does anyone have this amp with different N channel FETs? It doesn't make sense to have them so mismatched, almost as if they ran out of 28M15s and subbed in the 62N15P to finish a production run. What am I missing here? Thanks for any help!
Do you have a photo of the amp and or board number? Trying to figure out which P300-2 you have to be able to answer your question.
The amps I have here use FQA36P15 and FQA28N15 never seen or heard of the transistor you have in that amp. When I looked it up I could even find anything in the 247 package?
Yea, its is a IXYS FET, part number IXTQ62N15P. if you want to look it up. I have some 28N15s, going to try them if no one else chimes in. Just seems like a mismatch or a "throw this in there to finish the production run" kinda thing.
Maybe its a newer production than I have repaired. I have a few of the same board here and all are the above mentioned combo 36p and 28n.
You may wait for someone else to chime in. What is the number of the a/b driver card used in this amp?
You may wait for someone else to chime in. What is the number of the a/b driver card used in this amp?
Punch A/B gain card 1325-53110-02
Eq card 1325-52771-04
DTM2 P/S card 1325-54828-04
1 EMI shield card
and 2 output driver cards 1325-53617-01
Eq card 1325-52771-04
DTM2 P/S card 1325-54828-04
1 EMI shield card
and 2 output driver cards 1325-53617-01
My opinion... Unless there are design flaws that make an amp unreliable under normal operating conditions, use the same parts as were in it originally if they're still available.
While I 100% agree with Perry, and would normally order the correct parts, a minimum order of 30 with a 6 week lead time is no good (at over $3.15 per FET). That being said, I stuck a 28N15 in and ran some tests. Outputs from both channels were the same both on the scope and audible. Cranking up the volume, I used my temperature probe and checked the channel with the original parts, about 100 degrees F, the channel with the 28N15 ran at 111 degrees F. Looking at this amp, the only thing that seems to set it apart from the 200-2 (this is a 300-2) is the 62N15P, my thinking is that Rockford upgraded to the 28N15 to get just a bit more out of the amp (100W) and tweeked the circuit for it. even at idle the 28N15 side runs warmer. Any suggestions on a FET that will be above the 28N15 that is in stock at Mouser? Going to look myself, but opinions and advice always valued!
Did you readjust the bias after replacing the FETs? There's no reason that the 28n should run hotter at idle.
Also be aware that the heatsink isn't symmetrical which will make a difference between channels.
The 46N was used in many amps in place of the 28N. Some of the amps with the 46N used lower value gate resistors than the 28N amps (22 ohm to 10 ohms, IIRC).
fqa46n15 — Octopart
Also be aware that the heatsink isn't symmetrical which will make a difference between channels.
The 46N was used in many amps in place of the 28N. Some of the amps with the 46N used lower value gate resistors than the 28N amps (22 ohm to 10 ohms, IIRC).
fqa46n15 — Octopart
No I did not readjust the bias, as it was just a lop-sided test with the left channel having the 28N15 and the right channel having the factory 62N15P. Yes the heatsink is not symmetrical, but the left channel (the shorter side, not the back) also has the temp sensor, which should put the amp in thermal protection if needed. What is the full part number for the 46N? This amp has 10 Ohm gate resistors so that could be part of it. I would not be afraid of ordering the 46N part and trying it after adjusting the bias. That being said, I have noticed that in the RF amps I have repaired, with an exact swap of all new output FETs, I could never get them to overheat while leaving the factory bias setting. BTW Perry, can't thank you enough for your guide, posts, and overall being the superman amp repair guy!
The link I provided has the part number.
Could not get them to overheat at idle, or during bench-testing/burn-in?
Could not get them to overheat at idle, or during bench-testing/burn-in?
Crappers, didn't see the link Perry, it's 11 here and my bedtime is 8, gotta be up at 4 in the morning for work and I am a bit tired. BUT... the forums have been hopping and now the wife is up and giving me that look.. better pick all this back up in the morning. Other RF amps could not get them to overheat during full power bench testing. This one was just at idle on the temperature readings and then hooked to a DVC 4 ohm sub.
When I test, in 15 minutes, the amp is FAR too hot to hold my hand on. This applies to all but the class D amps and some of those will get that hot but it will take longer than 15 minutes.
Yes, during full power testing my aluminum encased load resistors, you could fry an egg on, and the amp, well I try and get my wife to touch them, but she just gives me the finger. They warp my little plastic sheets I use to protect the amps from scratches. I did not push this amp that hard, just doing a little "research" on what would happen if I stuck a 28N15 in the board.
FQA46N15 is the full part number.
In the future always reset the bias after any repair to the channel.
Confirm the gate resistors and drivers are the same for the 28N amps and the 62N amps.
In the future always reset the bias after any repair to the channel.
Confirm the gate resistors and drivers are the same for the 28N amps and the 62N amps.
Fixed! I got in some 46N15 FETs, set the bias on each channel to 50mA and ran her hard. The amp heated up nicely withing about 2-3 degrees on each FET. Got it cooking on the bench now for endurance testing. After that its good to go, thanks everyone.
- Status
- Not open for further replies.