loud hum w/ 1st Gainclone

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Well, I finished my 1st GC using BrianGT's boards/parts. I hooked it up to the computer output & turned it on. The output of the rectifier board was right on the money. I hooked up a single free air speaker (8 Ohm) to the right chanel and got a really loud hum--I was quite a bit put off, but had read about hums & thought I'd see if the hum would be present with a signal running through it. The output was ~2.2 VAC at the speaker. I opened up iTunes & turned on some music with no output & then increased volume. The net result was a blown fuse (2A) and an acrid smell from the speaker.

Well, before I replaced the 2A with another or larger fuse, I thought I question the collective wisdom about this. I traced lines & everything looks according to BrianGT's diagrams. To get a real clean look, however, I have to pull out the frame that the amp/rectifier board are mounted on. The box I built is nice looking, but a pain to work with. Any thoughts?

TIA,
Bret Morrow

PS I have 2 pictures of the guts, but I'm not sure of the value as the boards are difficult to see below the wire "cloud".
 
boholm said:
If you don't have anything connected to the input, does it hum then?

Yep, just as loudly. I haven't tried with music playing yet. I switched to a slo-blow so I'm ready to give it a try soon.

demogorgon said:
Is it the LM3875 or the LM4780?

a friend of mine has some problems with the lm4780 board, though his are just plain dead. all four of them..

regards
marius

Sorry, the LM3875 premium package.

Thanks & still looking for the answer!

Cheers,
Bret
 
JazzzSpazzz said:
Yep, just as loudly. I haven't tried with music playing yet. I switched to a slo-blow so I'm ready to give it a try soon.

Then there is no need to try with music for now. Not before the hum is gone. I do not know the boards, and I certainly do not know, how you have assembled it all. But there is definitely an error in the assembly - a loose wire, bad soldering, wrong wiring, whatever.

If you could make a sketch of how you have assembled it, it would help. And maybe even some pictures.
 
Nuuk said:
And measure the DC offset on the speaker terminals! ;)

Sorry, a bit of confusion on this side of the Atlantic. Isn't the signal from the amp to speakers AC? I did measure VAC at a steady 2.2 VAC--it fluxed up to 16VAC briefly when turning on. The "instrument" used to measure was a cheap digital multimeter. A standing 2.2 VAC differential across the + & - terminals of a speaker doesn't seem very health. I'm pulling the frame out tonight so I can carefully verify the wiring--this is likely where I screwed up. But my question is--should I be measuring VDC? javascript:smilie(':xeye:')

Thanks,
Bret Morrow
 
Bret, the signal to the speaker is AC but that's what is supposed to go to the speaker and won't do any harm at 'proper' levels.

DC is what destroys speakers and we need to keep it to a minimum to protect the speakers.

So measure the DC offset on the output of each channel with a 10 ohm resistor connected across them. You should be aiming for sub 100mV and the closer to 0mV the better. ;)
 
Ohhhhhhh

Nuuk said:
Bret, (SNIP) DC is what destroys speakers and we need to keep it to a minimum to protect the speakers.

So measure the DC offset on the output of each channel with a 10 ohm resistor connected across them. You should be aiming for sub 100mV and the closer to 0mV the better. ;)


Then DCV on the humming channel is 35V--I think that this may not be good ;-) The other channel comes in at 13.4mV--I have not listen to anything through this channel. I looked for misplaced wires, but everything looks good. Nothing seems loose either. I am going to take the amp board out of the box later tonight to look at the bottom side of the board to see if there is a short there. I have to pretend that I never read your site--esp. where you advise buillding the amp on a piece of wood before placing it into the box--I wish I would have done that first. Does the symptom (35VDC) indicate anything specific to you? (other than something is wrong ;-)

Thanks,
Bret Morrow
 
First, thanks for responding--I really appreciate the help!

Well, it seems that the bottom of the board looks clean--no shorts are visible. The board is BrianGT's and the V- & IN & OUT holes are separated--I also checked V+ & IN & OUT--no visible connections. So it doesn't appear to be shorted. The voltage is actually -35VDC not +35--sorry, my bad typing.

Nuuk, I did continuity checks with the power off between the V-, V+, PG-, & PG+ of the right (bad) channel to the left (good) channel. It showed 0 Ohms for each set (e.g. V- of Right to V- of Left). I think this should have checked for any break in the power rails & it looks like they are intact.

Any other test to look at? I've gone over the wires about 50 time & they are fine. I can't see a short on the board, but I'm at a loss to figure out why they huge DC offset. It seems as if it has to be a short between V- & In or OUT. The chip is not even the slightest bit warm, but I am only turning it on for a minute or so to test. Could the chip just be bad?

TIA,
Bret Morrow
 
JazzzSpazzz said:
Could the chip just be bad?

Since you have tested everything thoro'ly, it does seem that the chip is dead. So:

You should try and solder out the bad chip and measure it for a short between -V and output. If this is 0 ohms then it is dead. If not then measure between -V and output on the board. If no short there, then measure the input vs. -V. If no short there then power up the board(still without the chip) and measure voltage on the input pin. If this is not 0 volts, then measure +V and -V. If they are not + and - 35 volts, then . . . :confused: Well, the chip can still be bad, so a spare could come in handy ;)
 
My 0.2$'s:

In my haste to finish my first GC, I managed to mix up the signal and signal ground (the IN on Brian's board looks like an N - my anxious little brain said SG - signal, N - neutral, which makes no sense, but hey...).

This produced a nasty DC offset/buzz, but I can't remember if it was as high as 35V.

I feel your pain... But it's well worth the effort.
 
Nuuk said:
Do as Bo says but before you remove the chip, measure the voltages on each of the pins to check that you have both + and - 35 volts! ;)

Will do. I disassembled the unit holding the amp boards & got them out of the box with wires attached. Man, was this a bad design--it looks really nice (IMO), but it is not user friendly! I'm going to bring home some narrow test probes that I have at work before I test, so I will test tommorrow (mainly because I'm tired & don't want to make a mistake). Oh, dispite having electronic test equipment at work & am rather clueless about it--I'm better on the biology end

m@ said:
My 0.2$'s:

In my haste to finish my first GC, I managed to mix up the signal and signal ground (the IN on Brian's board looks like an N - my anxious little brain said SG - signal, N - neutral, which makes no sense, but hey...).


This produced a nasty DC offset/buzz, but I can't remember if it was as high as 35V.

I feel your pain... But it's well worth the effort.

Thanks, I check & rechecked, but I can always find other ways to mess things up ;-) I have to remember the end product right now!

Cheers,
Bret Morrow
 
Nuuk said:


I still maintain that hardwiring is the best way to build a GC for beginners as it is easier to fault find, and see (and understand) where all the parts go!

I'm not saying that PCB's are no good, just in this case I would prefer to hardwire. ;)

Sorry, I was saying my design of the box BrianGT's boards were put in was the "bad design". I think the boards were simple & quick. I read your site before Brian set up the group buy and was going to get around to trying to build 1 "someday". The availablity of a "kit" made me actually get going on it sooner, rather than later (never?). I have hardwired speaker crossovers with no concern but those components are huge compared to the ones in a GC.

My box is too small. The boards & heatsink are on a platform that drops into the box--it is very snug & I didn't think enough about soldering lines from the tranny & input/outputs when I set it up. The goal was to use silicone adhevsive to secure the platform to small cleats on the sides of the box when everything works--I thought that it would make a real nice dampening system for the amp boards. It took me a good hour to get the platform out of the box, so I could remove the right side board to see its underside. The best thing I can say about it is that I still think that the box looks lovely on the outside and should work well, eventually.

I hope to test out the chip this PM.

Thanks again,
Bret
 
results from testing

Well, I measured to try & discover the problem, but I'm still a bit clueless. Here's what I know:

No continuity from chip pin 4 to pin 3 (-V to OUT).
No continuity from chip pin 4 to OUT on board.
No continuity from chip pin 4 to pin 7 (-V to IN).

I powered up the board & measured VDC:
...from V+ to PG+ I've got +35VDC
...from V- to PG- I've got -35VDC
...from pin 4 to board ground I've got -35VDC
...from IN to SG I've got 10.3VDC
& of course I've got -35VDC from pin 3 to OG.

The only thing that I can identify as odd is the chip, but there appears to be no apparent shorts in the chip. I emailed BrianGT telling him what I know & asking him what he thinks.

I'm thinking about desoldering the leads & listening to the left channel in mono! Stereo is so over-rated!

Cheers,
Bret Morrow
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.