Trouble with 64-step relay attenuator (RelVol1)

[Ralph]

I am building an all new pre amplifier combining the P06a pre amp from Rod Elliott and the 100k RelVol1 kit from Dantimax. It's a 64 step relay attenuator with 6 relays.

The problem is that on some steps the sound gets way louder than it should be. The next step is back to normal levels. It happens every 8th step from 0-63 (7, 15, 23, 31, 39 etc). This is when the first 3 relays (1, 2, 4) are always switched in.

What did I check:
- I checked all resistor values, they are as specified (no mixup).
- Also I checked if the values are correct for a 64-step 1.5dB per step setup. Seems ok (see worksheet 3)
- Resolderd every joint.
- All relays switch on/off when they should be.

Is there some other thing I can check? I measured resistance per step and pre amp output (ref 1V in) per step, please take a look at the Excel spreadsheet (the yellow steps are the problem ones). Any help is much appreciated!

http://www.jonkers.net/audio/relvol1.xls

Ralph

[gootee]

Can you post a schematic, or a link to the schematic?

[Ralph]

This is a link to the schematic. http://electronics.dantimax.dk/SiDow...elVol1_sch.gif

As you can see, R1..R24 are the resistors that determine the steps. I have verified that they are of the correct values and that I soldered them correctly.

R27/28 are the loading resistors. For a 100K load I use 11K1 resistors for each.

After doing some more testing Today, I determined that if I remove the loading resistors all together, the steps are working fine! There is no more jumping in level. However, from steps 10 or so down, the level increases a bit and is very thin sounding. So obviously it's still not entirely correct. But I'm guessing that the loading resistors are the cause of the problem.

Ralph

[Ralph]

Problem fixed I think. I was using incorrect loading resistors (11K2). This also gives me the problems in the simulation. I'll replace them with 100K ones tonight and I should be fine.