Repairing a Hifonics BXi2006
I'm repairing a Hifonics BXi2006D.
I was running it at 2 ohms a few weeks ago when it released the magic blue smoke. The funny thing was I wasn't being particularly abusive, I had it turned down so that I could talk to my friends as we drove.
Anyways one half of the PSU is blown, a couple of the FETs are visibly damaged. I intend on replacing them all, since there could be some latent damage that'll show up later.
As for replacing them, they are FPQ85N06 FETs.
Using the filter tools on Digikey I found the AOT460 as a potential replacement FET. It appears to have a lower RDSon and gate charge and higher power disipation, all of which are desirable. (At least I'm pretty sure that they are desirable). all at a cost that's about 25% less than the original parts. The main thing I'm not sure about is the specs are given with ID=30A for RDSon, whereas it's at 60 something amps for the 10mOhm RDSon of the IRF85N06. So I'm not sure if I can even compare the two.
The other thing I'm not sure about is the AOT460 being a logic level FET. I made sure the gate charge was acceptable at 10V, which is what the driver would be using. The VGS is 20V so it can handle the higher than logic level drive. I'm thinking this just means it'll be in the linear region for less time than the original part. Which I think is a good thing.
So I'm hoping that this means I can get a better part at a lower cost. If I'm heading down the wrong path someone please let me know.
Also if anyone has any idea of what would cause the amp to blow I'd love to hear it. There was 0 abuse, the chassis wasn't even warm. I'm perplexed.
Digi-Key Part Search
Digi-Key - FQP85N06-ND (Manufacturer - FQP85N06)
I'd use the IRF3205 if you don't want to use the originals.
Logic level FETs can cause a problem because the turn-off drive signal produced by the drive circuit doesn't go instantly to ground. The slope at the bottom of the waveform is acceptable to normal FETs but logic level FETs will sometimes have cross-conduction which will cause the FETs to fail.
If the power supply is regulated, the problem may go unnoticed because the low duty cycle at idle will allow the FETs to turn off before the other half turn on. Only when driven hard will they have problems but the additional current draw due to cross-conduction won't be noticed because the amp would already be drawing significant current.
If you use the 3205s, and the gate resistors are greater than 47 ohms, I'd suggest replacing them with 47 ohm resistors.
Thanks for the response, Perry. You were just who I was hoping for.
The stock gate resistors of this amp are 22 ohms, so I think I'll just stick with that.
I'm 90% sure the amp is unregulated. This might be a dumb question, but with a typical push pull power supply, where the center tap of the torroid is connected to Vbat and the other two legs are switched to ground by the 2 FET banks, can there really be cross conduction by the FETs? I don't think it's set up like a bridge where there would be essentially a dead short from the positive to negative rail. Or is the issue that the logic level FETs would stay in the linear region when the drive signal goes low?
Either way I'll stay away from the logic level FETs, thanks for clearing that up for me. Hopefully I'm not coming off too annoying. I guess what I'm after isn't what a good replacement part is, so much as what makes it a good replacement. For now (until I understand the situation a bit better) I want to stay away from the 3205s, since they cost roughly the same as the original parts. It seems like I can get suitable replacements for cheaper. "Seems" being key here. My understanding of FETs isn't as good as it could be, in school we really just glossed over them.
For example, I found these FETs. The only spec on them that is worse is the the power max spec of 149W instead of 160W. (6.875%) That said the RDSon is lower (22%) and the gate charge is lower(53%), so I'm expecting the power dissipation to be lower as well. So I think I should be able to get away with less power handling. And they're fairly cheap compared to the original part. Maybe there's a good reason for that. But to me it seems like these would work well.
Digi-Key - 568-4970-5-ND (Manufacturer - PSMN7R6-60PS,127)
IRF3205PBF for International Rectifier | Arrow Electronics Components
$0.56 each for 3205s.
Sometimes it's difficult to determine what's a good sub by looking at the specs. Power supplies are more forgiving than class D (or even class AB) audio but it's best to use parts that are proven to be reliable.
Thanks for the advice Perry. I appreciate it.
Oh and thanks for the link, that's considerably cheaper than digikey.
|All times are GMT. The time now is 05:27 AM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright ©1999-2017 diyAudio