| Nappylady |
LM3875. Single chip amplifier. +/-38 volt rails, 1000 uF filters. I'm 99.9% sure I got everything right--nothing blew up, and my meter told me all was well.
Still, I hooked up a speaker and--+V rail appears at the output. Whoops.
Does this mean the chip is dead? Or do I just have something wired wrong? It's not a super big deal... it was an ugly hackjob anyway... but from what I'd read, the chips mostly didn't swing one way or the other when they died... This is my first attempt and I don't know if I should just move on or check my connections for a 43rd time...
Thanks all.
(PS - it actually made a pretty neat noise when I turned it on with a load connected... too bad it burned out the speaker in the process. :bawling: ) |
|
|
| li_gangyi |
| check your other rail...might be disconnected or your input might be tied somewhere...where it's not suppossed to be~!!! |
|
|
| lieven |
Been there, seen that....
Please explain first your setup to us (inverting igc or non-inverting nigc) + resistor and cap values.:scratch: |
|
|
| antomas |
Hi Nappylady,
+-38V seems to me a bit too close to the voltage limit of +-42V (datasheet page 3). Could this be the cause of your problem?
Ciao, Massimo |
|
|
| Matttcattt |
| this happened to me, with my OPA541 amp, turned out, i connected +v and -v the wrong way round on both amps, very confusing diagram (OPA541AM). |
|
|
| Stabist |
Hi!
A friend of mine and I also killed one of the GC yesterday ... :(
Pitty ...
Still - one picture of the "protoype" that didn't survive the burning in .... |
|
|
| Stabist |
But than again - it might be the destiny - 'cause now the chassis will be redesigned a bit ... And also the topology will be changed ...
Hush .. Still a bit sad because of poor LM's that got fried in an unfriendly maner ;)
And another picture ... |
|
|
| jackinnj |
GC's don't like to be sine-wave tested (although we don't know from your spec if this was what you were up to.)
Further, you really have to pay attention to the heat sink specs, hf oscillation etc. These big op-amps are not quite that simple to implement. |
|
|
| ronc |
I love the smell of fried silicone in the morning.
ron |
|
|
| jackinnj |
| quote: | Originally posted by ron clarke
I love the smell of fried silicone in the morning.
ron |
I prefer bacon, resistors and coffee |
|
|
| li_gangyi |
| nawz...I prefer Oil filled caps...filled with PCBs...and stuff...and selenium rectifiers... |
|
|
| Nappylady |
Details, details...
It's the most sloppy, horrible implementation ever. I'm positive that I got the connections right--the problem was, I'm sure, my own laziness.
One question that I didn't ask, out of both laziness and embarassment, because at the time I bought my caps, I thought I'd be using a 35 volt rail... just how bad is it to run these things outside spec? They've gotta have some headroom, right? 38V isn't THAT much above the rating...
...
The caps are just fine. No signs of degradation. There isn't actually any sign of trouble anywhere, other than the whole project looks like $#!%. The only thing that's amiss is that, while I was testing, I did a boo-boo with a multimeter and momentarily shorted one of the transformer leads to ground... it sparked alot, gave my dad a heart attack, but didn't even throw the breaker.
I'm moving on to a better project now, with proper circuit boards and stuff... this time I AM going to trim off the unused leads on the chip and I AM going to get capacitors the right rating... and I'm going to ground everything properly and take my time.
Still, I can't help wondering... is my chip maybe still good? Is it worth the trouble of pulling it out of its current home or should I just toss the little POS PCB I used?
Thanks everyone... |
|
|
| li_gangyi |
| try another chip...and be careful where you probe arund... |
|
|
| usekgb |
Hey Nappy,
I'm in Salt Lake too and am working on a Gainclone right now. We should put our heads together and figure these things out.
Cheers,
Zach |
|
|
| li_gangyi |
| I think GCs are really easy to built...as long as you watch you rails and stuff...should be ok after that... |
|
|
| ronc |
Rails and grounds.rails and grounds.
I have fried chips but it was my own dumb mistakes.
The voltage that nappylady is running at sounds a bit high for me.
ron |
|
|
| usekgb |
When I first slapped my gainclone together, it just wouldn't pass audio. I had 2 or 3 mV DC offset at the output, but no audio. It just seemed really wierd. It could be because I just grabbed parts out of my spare parts bin, but gainclones are supposed to be pretty tolerant of different parts. It should have at least passed audio.
Nappy, if your chip is fried, I've got 2 more at the shop. Let me know if you need another one.
Cheers,
Zach |
|
|
| li_gangyi |
| Guess being careful pays here... |
|
|
| antomas |
| quote: | | One question that I didn't ask, out of both laziness and embarassment, because at the time I bought my caps, I thought I'd be using a 35 volt rail... just how bad is it to run these things outside spec? They've gotta have some headroom, right? 38V isn't THAT much above the rating... |
Hi,
if you browse around this forum, you will find many Gainclone implementations using:
2x18V/160VA (and at least 1000uF/35V caps) on a 4 Ohms load
2x24V/160VA (and at least 1000uF/50V caps) on a 8 Ohms load
These values should be ok to prevent the current limiter of the 3875 from switching on and to have a reasonable dissipation of power (=less heat).
Ciao, Massimo |
|
|
| Nappylady |
UseKQB: I'd be very interested in seeing your work in person; I've been dying to get out and don't have a car. I agree--we should put our heads together.
I have 4 more chips and I plan on ordering some more from National--eventually I want to have a 6-channel amp to replace the one I blew up last month. (That was an RCA home-theatre amp that, after opening up its poor fried carcass, found out it was a hybrid--no wonder it sounded so bad--and the thermal grease hadn't even turned brown. Those pioneer chips apparently have a glass jaw...)
I bought one of those nifty copper-clad fiberglass PC boards from Radio Shack and I'm working today on the layout and traces and stuff. Please email me at dr.napalm@attbi.com and we can talk in more detail... |
|
|
| jackinnj |
| here's a dead one -- too much sine wave testing -- this puppy was happy putting out almost 50W at 0.011% THD, then went poof. Note the blown V- capacitor (1000u/63V). |
|
|
| li_gangyi |
| ...how did you blow that 63V cap?? too high inpout voltage..should it blow?? I dunno I thought sine-wave was the best waveform for testing... |
|
|
| matjans |
did you really blow that cap with sine-waves? I was about to borrow a scope & freq. generator but hey....
I really can't imagine blowing a cap that way ...
How did you test it? What frequencies did you test on ? |
|
|
| ir |
| are you very sure the cap's not in backwards? a 63V cap should 'survive' to about 80V for a short burst, however if it's polarity was reversed, death could take up to a minute or so as the electrolyte heats up eventually boiling (then again, death could be instant too) |
|
|
| li_gangyi |
| I'll bet it was reversed...no way you can kill a 63V cap liddat... |
|
|
| SkinnyBoy |
| well.. if the dark we can see through the board is tracks, why would we be able to see the tracks throught the board? hmm.. anyway.. if they are.. then either both capacitors are in the right way.. or neither are... |
|
|
| jackinnj |
uuuuhhhhhh, I think you guys are right. I had some boards made with a different capacitor orientation (perpendicular to that shown) and just soldered them up.
They make a nice loud popping noise. |
|
|
| SkinnyBoy |
| quote: | Originally posted by jackinnj
uuuuhhhhhh, I think you guys are right. I had some boards made with a different capacitor orientation (perpendicular to that shown) and just soldered them up.
They make a nice loud popping noise. |
:rolleyes: :rolleyes: :rolleyes: :dead: :bawling: :dead: :bawling: :bawling: :rolleyes: :att'n: :bawling: :hot: :xeye: :rolleyes: :hot: :att'n: |
|
|
| li_gangyi |
| told you...reversed caps...are BAD |
|
|
|
