If you were to guess what I killed on a F5 when the output shorted?

[xferboy]

Well,

I've been lurking here for a while now. A friend of mine got me in on a set of DIY F5's and F4's. Got the F5's assembled, biased, all peachy, even tested each channel against a speaker (would have done both at the same time, but found out I had a dead lowther )

anyway, long story short, after I found a set of used Heresy's I wanted to re hook everything up and listen.

I'm pretty sure I shorted a speaker lead on one channel and let some of the magic smoke out...

Now with the speakers and inputs disconnected it takes about 10 seconds after powering on for wifs of smoke to start appear (when the speakers were hooked up, there was more than a wif!!). I can't tell exactly where it is coming from.

Little disappointed in myself, so I'll leave them sit for a couple of days (busy weekend coming up anyway).

If any of the smart folk out there were to guess (or know from experience), which component(s) would you guess were baked? Oh, by the way, I now have 24v DC across the output from the bad channel. I tested before hooking out the inputs and outputs and it was ~0.02 v across the output.

Thoughts, comments????

thanks

Allan

[d to the g]

I'm no expert, but I'd guess that all of the fets/jfets are toast, including the current limiting bipolars. I'd also check all of your resistors with a dmm and make sure that they read correctly (with the fets off the board).

[AndrewT]

I suspect some of the FETs are shorted D to S.
I also suspect that the remaining FETs have been operated outside their SOA.
Change everthing.

Was the current limiter fitted?

Always use a bulb tester when ever a project is powered up after modifying. ALWAYS.

Never hot swap plugs and sockets, as it can lead to tragedy.

[xferboy]

Thanks of the input!!

I did a little looking this morning.

Did some resistance measurements between the good and the bad and most values are close.

Then I did some quick testing with out desoldering (just some comparison testing) around the semi's and using the diode test this is where I found differences. The big stuff measured the same between good and bad, but around the little semi's, there were differences. I'll have to get some replacements ordered and swap all of them.

My friend bought all the parts, otherwise, I would have ordered spares.

AndrewT: Bulb tester... is this just a standard light bulb in series with the AC? I looked at some of your other posts and that's what I can gather, but just want to confirm.... what kind of wattage?

Thanks

Allan

[juma]

Originally Posted by xferboy

.... it takes about 10 seconds after powering on for wifs of smoke to start appear...

That's the time needed to heat power resistor enough to make them smoke. Check for changes of colour on them. If you spot "heat stains" on 0R5 resistors it means that too high current runs through them - so either MOSFETs are toasted (D-S short as Andrew said) or they have too high VGS.
Now, D-S short is likely to burn those resistors in a couple of seconds, but you have repeating process of smoking which leads me to belive that MOSFETs are still alive but with too high Vgs - maybe 4k7 NTCs are mounted so that they are touching something and causing too high Vgs ?
Check carefully for possible shorts in circuit, loose wires, small pieces of metal....

[stefan5000]

One of the power mosfets gone along with the 0,47 resistor!

[AndrewT]

Originally Posted by xferboy

Bulb tester... is this just a standard light bulb in series with the AC? I looked at some of your other posts and that's what I can gather, but just want to confirm.... what kind of wattage?

ClassAB can start with 40W and will need higher if the Ib is big.
ClassA cannot be started with low wattage bulbs.
You will probably find that 150W will not let it bias properly and when the bulb is that big it won't be as effective at preventing damage.
If your ClassA bias is turned right down, you can start it just like any ClassAB.

[tinitus]

Originally Posted by AndrewT

If your ClassA bias is turned right down, you can start it just like any ClassAB.

[xferboy]

little semi's pulled out, 2 of 4 have shorts (compared to the other 2), parts on order, light tester ('safety light') built . never heard of doing this before, great idea!

thanks

Allan

[woodturner-fran]

When I built mine, I had a short from the fets to the heatsink (long story). Each time it took out the 0.47R in the space of seconds - I didn;t notice for ages that it was gone. Almost no discolouration of the resistor etc etc. The fet itself was AOK - I think something else would blow before those - Aren't they rated for 20A current and nearly have to glow white before dying? Sounds like you have the problem diagnosed though....


Fran