Another first gaincloner...

[toque_101]

Been listening to my first gainclone for a few days now. I love it! Sounds great (in my non-audiophile opinion). Its based on the lm3886, used Elna 1000uF filter caps and Solen 4.7uF for coupling. Standard resistors, nothing special here (22k / 1k for feedback, 22k shunt). I have 5600uF in the power supply in parallel with the Elna's, going to try removing them and seeing what happens there, just don't want to power it down right now I accidently left it on when I left the house today (don't really trust my PSU wiring yet), sounds better I think now that its "warmed up" for the day. For a schematic, check out GregGC's semi-ultimate gainclone:
http://www.diyaudio.com/forums/showt...500#post259500

Exceptions are:
- no POT on the input (use my soundcard for volume)
- R5 = 1k
- 5600uF in parallel with the 1000uF
- LM3886TF instead

I have a slight (what I think is 60Hz) hum in both channels (could be because my torroid is close to the chips w/ no sheilding really). Also, there is a higher pitched "hiss" from the tweeters at idle. In both aspects, removing the source doesn't change it. I figure this is a grounding problem. I did two separate star grounds (signal and power) for each channel and then connected them with a single wire. Also, I pick up a bit of a radio station on the left channel, basically have to glue my ear to the tweeter to hear it.

I'm going to try to attach some pics, we'll see if they come through.

[toque_101]

.

[toque_101]

bad pic of the PSU

[b<ben>n]

looking good.
im gona make a amp with that chip.

[Morse]

Hi Toque101;

Are you *sure* it's a 60cps hum? Check and make sure it's not 120cps - if so, it's not the PS trafo, but the rectified output from it that's inductively coupling with the signal input leads. Suggested attempts at a cure would be to put EMI bulkheads around the signal inputs as close as possible to the pins. Another suggestion would be to use twisted pairs or shield the V+/V- leads (sorry, can't tell if you've already done this from the photos - combo of a poor quality monitor and worse quality eyesight over here).

Good luck and all the best!
Morse

[Nuuk]

The leadout from the Solen cap on the right runs VERY close to the metal case

[toque_101]

Originally posted by Nuuk
The leadout from the Solen cap on the right runs VERY close to the metal case

Its looks almost like its touching in the picture, but its actually about 1/4" from touching. Are you saying that I could be picking up EMI from it?

[toque_101]

Originally posted by Morse
Hi Toque101;

Are you *sure* it's a 60cps hum? Check and make sure it's not 120cps - if so, it's not the PS trafo, but the rectified output from it that's inductively coupling with the signal input leads. Suggested attempts at a cure would be to put EMI bulkheads around the signal inputs as close as possible to the pins. Another suggestion would be to use twisted pairs or shield the V+/V- leads (sorry, can't tell if you've already done this from the photos - combo of a poor quality monitor and worse quality eyesight over here).

Good luck and all the best!
Morse

I'm not ~sure~ its 60Hz, actually, its probably more likely to be 120Hz because the response of my speakers only goes down to 100Hz.

EMI bulkheads? How would I do this? Examples I can look at anywhere?

Use twisted pair for the power cables? So, something like CAT5 network cable then?

Oh yeah, that radio station stuff I was hearing seems to be the fault of my source cable. When I position it differently I get a stronger or a weaker radio signal depending. I guess I should invest in a better sheilded cable. I should probably be glad I'm only getting the noise I am, considering I have 2 computers, a monitor and a florescent light on the same desk as my amp

Thanks Morse and NUUK for your help.

[GregGC]

Originally posted by toque_101


I'm not ~sure~ its 60Hz, actually, its probably more likely to be 120Hz because the response of my speakers only goes down to 100Hz.

EMI bulkheads? How would I do this? Examples I can look at anywhere?

Use twisted pair for the power cables? So, something like CAT5 network cable then?

Oh yeah, that radio station stuff I was hearing seems to be the fault of my source cable. When I position it differently I get a stronger or a weaker radio signal depending. I guess I should invest in a better sheilded cable. I should probably be glad I'm only getting the noise I am, considering I have 2 computers, a monitor and a florescent light on the same desk as my amp

Thanks Morse and NUUK for your help.

Nice work so far.

As far as the RFI I don't think more money spent on cable will do it.

First you really have to shield the whole amp (close the lid and see if it's still the same). Secund I'd use 470R/1nF LP filter right at the RCA (before the input cap). Like here (R1/C1) but with values I suggested above:
http://www.diyaudio.com/forums/showt...481#post309481
Here it is in the case:
http://www.diyaudio.com/forums/showt...939#post272939
It wont cost you anything and it'll solve the problem if you have really strong RFI. Lastly (or maybe first I'd check the wiring again). I can't understand where is the input cap connected too. It looks like it's connected to the last pin of the IC. Also there is a black wire (22k resistor I think) that goes to the GND-star. I'd connect it to the same GND that the 1k resistor is connected to or connect the 1k res to the same gnd point where the 22k res is connected to, not like in your case.

Hope it helps.

/Greg

[toque_101]

Originally posted by GregGC


Nice work so far.

As far as the RFI I don't think more money spent on cable will do it.

First you really have to shield the whole amp (close the lid and see if it's still the same). Secund I'd use 470R/1nF LP filter right at the RCA (before the input cap). Like here (R1/C1) but with values I suggested above:
http://www.diyaudio.com/forums/showt...481#post309481
Here it is in the case:
http://www.diyaudio.com/forums/showt...939#post272939
It wont cost you anything and it'll solve the problem if you have really strong RFI.

- Cool, I'll have to try the filter. I don't have any 470 ohm resistors kicking around so I'll have to get out and pick some up. I'll let you know when I get to it (ie: after the tons of homework, labs and assignments I have due before exams ).

Lastly (or maybe first I'd check the wiring again). I can't understand where is the input cap connected too. It looks like it's connected to the last pin of the IC. Also there is a black wire (22k resistor I think) that goes to the GND-star. I'd connect it to the same GND that the 1k resistor is connected to or connect the 1k res to the same gnd point where the 22k res is connected to, not like in your case.

Hope it helps.

/Greg

The leg of the input cap is sorta wedged between the 10th and 11th pins and soldered there. Pin 11 is a "NC" pin so basically I combined pins 11 and 10 so I could have more "real-estate" to solder to.

The 22K resistor and the 1K resistor are connected to the signal star ground (the ground connection on the RCA jack); also, the black wire serves to connect the two stars together (signal and power). Its hard to see the 22K in the picture because the black wire hides it a bit.

Thanks Greg!

-steve