Page 1 of 1

Mark III noise question

PostPosted: Sat Jun 29, 2013 7:24 pm
by Soundguy2856
I've just finished the first of 2 Mark III restores using SDS cap board and Poseidon driver board. First test was fine using a fair set of tubes I had on hand. Power is good....freq response is good....distortion is good ...basically all is as expected but I have about 30mV of residual hum in the noise floor. Is that to be expected? The freq of the residual is 60Hz which makes me think it's source is the half wave bias supply.....but maybe, with this combination, 30mV is normal? Thanks for your input.

Re: Mark III noise question

PostPosted: Sun Jun 30, 2013 5:21 am
by Shannon Parks
It should be under 1mV. The C- bias isn't the first place I'd look. Look at heater ground referencing and RCA input chassis isolation.

Shannon

Re: Mark III noise question

PostPosted: Tue Jul 09, 2013 2:25 pm
by Soundguy2856
Shannon/All - On your Poseidon board the ground point is configured as a star ground for the low level driver stuff. I took that point to my overall amp star ground point at a terminal strip on the chassis that was in reach of the transformer HV CT etc. I suppose I could have taken everything to the poseidon star ground. Was that the intention? I suppose I could have a circulating current form the low level board? I've definitely got the heater CT tied through the .02uf cap. Any thoughts?

Re: Mark III noise question

PostPosted: Sat Jul 13, 2013 6:00 am
by Shannon Parks
All the low level input stuff should connect to the rest of the amp from 'C' (to the main original star ground). So you if disconnect 'C', your driver board and RCA ground should be high impedance when measured to the rest of the chassis. So try doing that and measure with a DMM - you should be isolated.

Shannon

Re: Mark III noise question

PostPosted: Tue Jul 23, 2013 3:24 pm
by Soundguy2856
I disconnected point C and I get 2k8 to chassis which is the feedback resistor through the output tranny gound reference. In my current build all the items that require a gnd ref are taken via there own 18Ga wire to a central point which is also where my HV ct and filament ref cap, power cord 3rd wire & chassis all come together. The hum I get is about 20mV and at 60Hz which tells me it's a common path through something with a common impedance. I've separated all the cap sections on the SDS board and taken them back to star gnd individually....no difference. If I pull the tubes out of the low level board the hum goes way down and becomes 120Hz which makes sense becasue that would be ripple related. Are there any items that shoud be reference together perhaps at point C rather than taken back to star gnd individually? Not sure how the haphazard scheme of the original Mark iii performs better than this? Any help is appreciated.

Re: Mark III noise question

PostPosted: Tue Jul 23, 2013 5:42 pm
by Shannon Parks
Could you take some big pix and send them to me? Or post them? That noise level is ridiculous, so there's a big problem somewhere. Also, could you post your voltages per page 9? Resistances from page 10, too. That might give a clue.

Shannon

Re: Mark III noise question

PostPosted: Wed Jul 24, 2013 1:07 pm
by Soundguy2856
OK - Problem Solved -
Actually I believe that there were 2 problems...one is the common gound on the SDS board which ties all the cap stages together.....the charging current of the fisrt cap stage is imposed on the ground seen by all the others....but after separating and making that change I still had 60Hz hum. So lesson re-learned...start with the basics....Following Shannon's advice I started measuring voltages and pretty quickly realized that the first input stage of the low level board was drawing more than a normal share of current. Long story short the first stage 12AX7 was to blame......Bad old RCA tube! I changed it and went from this:
Image

to this:
Image

thanks and let the music play!