[Pars]

I'm building up a dynamic on a V2 board from Justin, and have it to the point where the amp section is more or less done. I started testing it this weekend, and while the offset is to a manageable level, I had a couple of questions.

First off, I am powering it with a B&K triple bench supply, using the two 0-25V outputs, wired + to - with B tracks A on, so +/-16V. Initially, I tied in the chassis ground, but have since removed that. It is just jumpered into the board with jumper cables. The L and R ins are jumpered to board ground, and since I don't have a pot in yet, I went ahead and populated the 500K input resistor in each channel with 475K resistors.

Offset is ~10mV R, 32mV L channel, without the servo opamps in. Initially, I did not match resistors, but have subsequently matched up the 500 ohm, 200 ohm and 5K resistors well within 0.5%, probably within 0.1% with a 6 1/2 digit HP3268 bench meter.

I ran across a strange problem, in that using a crap pair of Sony MDR-201 phones, connecting the right channel to either L or R output causes a rather bad hum, like a ground loop. Offset on both channels shoots up to 300-600+ mV, and the amp draws more current... about 125 mA per rail normal, up to about 150+ mA with these phones connected. L channel alone seems OK, although I can still hear a bit of hum. The amp works fine playing music, btw.

I tried out some Grado SR-60s, and these do hum, offset went up to 60-90 mV, so I pulled them off quickly.

Any thoughts? Is this because the amp is uncased, etc., or should I be looking for something else?

Also, suggestions on getting the non-servo offset down, particularly in the L channel? I have read most of the posts on this, with dip16amp and Garbz etc. I have plenty of 2SA1015s and 2SC1815s, but did not order any extra input FETs.

Thanks for any suggestions, as until the hum issue, I thought this one was going to be a piece of cake

.

Chris

 

[dip16amp]

Ground should be connected from the power supply to the board as well as ground from the headphones and ground from the source.
Changing LEDs should get the left channel offset down to less than 10 mV.
The pot also connects to ground when installed.

 

[Garbz]

see my headwize post for my answers.

I'd just like to add to be carefull with the signal ground and earth mains. My chassis is hooked to earth mains, and my intputs/ outputs float. The pot was originally connected to signal grounds however when it went into the chassis it made a connection to earth and shorted the two together. huummm.

 

[Pars]

Thanks for the replies.

Quote:

Originally Posted by dip16amp Ground should be connected from the power supply to the board as well as ground from the headphones and ground from the source. Changing LEDs should get the left channel offset down to less than 10 mV. The pot also connects to ground when installed.

Ground is connected, but I left the PS floating. + from channel B is V+, - of channel B and + of channel A are connected together and are the ground. - of channel A is V-. There is a green chassis ground banana connector that you can tie into the "ground" if you do want the ground referenced to the supply. All inputs/outputs are to signal ground (i..e., on board).

Going thru my 20 LEDs, I cannot get the left channel any lower than it is. I can make it go up (to about 60mV) however.

Thinking about it (the hum), and a thread here or on headwize that Kevin had with Jeffreyj, I think that this may relate to PSRR issues. It acts almost like it is going into oscillation. Unfortunately, I don't have the transformer and some other parts needed to complete the PSU onboard yet.

 

[Pars]

Quote:

Originally Posted by Pars Thinking about it (the hum), and a thread here or on headwize that Kevin had with Jeffreyj, I think that this may relate to PSRR issues. It acts almost like it is going into oscillation. Unfortunately, I don't have the transformer and some other parts needed to complete the PSU onboard yet.

Man, this is bad, quoting myself

I realize there areabout 90 Gilmore threads going on here right now, and that's a good thing. However, I was hoping for some input regarding what I am seeing Quote:

I ran across a strange problem, in that using a crap pair of Sony MDR-201 phones, connecting the right channel to either L or R output causes a rather bad hum, like a ground loop. Offset on both channels shoots up to 300-600+ mV, and the amp draws more current... about 125 mA per rail normal, up to about 150+ mA with these phones connected. L channel alone seems OK, although I can still hear a bit of hum. The amp works fine playing music, btw.

and I was hoping for any further suggestions as to fixing this? The amp is currently not connected to an earth ground thru the bench supply or a source (battery PCDP), so I can't imagine its a ground loop? Thanks for any suggestions. I did get the offset of the left channel down to <15mV no servo, so that is getting better at least. Both channels go to 2mV or less with the servo opamps in.