I've built one channel of a Gainclone (LM3886). At medium-high volume, the sound starts to crackle at a definate point, below which is fine. I tried putting a zobel network (100 nf and 10 ohms) across the output and it seemed to reduce the noise, but it can still be heard.
What else can be done to help?
Many thanks.
What else can be done to help?
Many thanks.
Sorry, been asleap.
Yeah, I have a large heatsink fitted to the chip. 4 x 2 inches. It gets slightly warm to the touch, so I don't think the chip is overheating.
The transformer is 160 va, and the power rails are +- ~38v. The caps are 5600 uf, and the amp is non-inverting. The speaker inpeedence is 8 ohms.
Many thanks.
Yeah, I have a large heatsink fitted to the chip. 4 x 2 inches. It gets slightly warm to the touch, so I don't think the chip is overheating.
The transformer is 160 va, and the power rails are +- ~38v. The caps are 5600 uf, and the amp is non-inverting. The speaker inpeedence is 8 ohms.
Many thanks.
Hi Mat,
have you eliminated air from the gap between heatsink and chipamp?
You want the gap as small as possible (0.0001inch=0.002mm) and FILLED with thermal transfer compound.
+-38Vdc is towards the top end of supply voltage for 8ohm speakers but if the correctly coupled heatsink is not getting beyond warm then it indicates that excess heat is not the problem.
Could it be oscillation brought about as output voltage hits a threshold? What would be the symptom? internal triggering on detection of excess HF?
My discrete Crimson has a thyristor lock out on detection of HF @ the VAS. It ONLY ever latched when pushing 50kHz at near full power
I didn't try that again
have you eliminated air from the gap between heatsink and chipamp?
You want the gap as small as possible (0.0001inch=0.002mm) and FILLED with thermal transfer compound.
+-38Vdc is towards the top end of supply voltage for 8ohm speakers but if the correctly coupled heatsink is not getting beyond warm then it indicates that excess heat is not the problem.
Could it be oscillation brought about as output voltage hits a threshold? What would be the symptom? internal triggering on detection of excess HF?
My discrete Crimson has a thyristor lock out on detection of HF @ the VAS. It ONLY ever latched when pushing 50kHz at near full power
I didn't try that againAndrewT said:
Or that the chip isn't making good contact with the heatsink? Maybe unlikely, but possible
EDIT: Ah, oops. Missed you asking whether the poster had got rid of the air gap with paste and all that!
I heard mine make a crackle noise once, but then it didn't seem to happen again, so I am quite curious as to what it is
I had a similar problem with my bi-amp system where all channels would crackle at moderate listening levels. Trouble shot that one for quite some time.
As it turned out the power switch was defective, arcing or something internal when I tried to run some current thru it. Now I use a power strip to do that chore. Problem solved.
Hope this helps.
David
As it turned out the power switch was defective, arcing or something internal when I tried to run some current thru it. Now I use a power strip to do that chore. Problem solved.
Hope this helps.
David
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.