Adding a little to the lead resistor R7 does one trick. The second is to decrease R8, the 3.3K and add a few hundred pFs to C1, or at least so says the sim. Will have to wait until Friday.
An externally hosted image should be here but it was not working when we last tested it.
Well, at this point, you're +/-0.2dB, which ain't bad. Channel to channel matching is the most important and your data looks pretty darn tight. I did things the hard way, lots of trimming, but I don't have an inverse RIAA that I have confidence (read: calibration) in. I'll probably pull the network components off my perf board to stuff in the PCBs, just to save a few steps.
Some channel to channel matching --
An externally hosted image should be here but it was not working when we last tested it.
I was reading the Sowter 8055 datasheet -- it's been changed a bit to include some recommendations for a Zobel. The Denon cartridge I use has a 40 ohm impedance, so I added 20 ohms to the output of my generator (it has 20 ohms output impedance). I used a 740pF variable cab and 2.4K resistor (in parallel with the 6.8K) to look at the input at the grid of the D3a. The 190pF +2.4K reduces the bandwidth 0.4dB @22kHz.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Attachments
here are some more charts -- i took a look at the frequency response with the zobel -- using a 40-740pF variable cap and 2.4k, and 3.18uS compensation:
and a comparison of the different D3a's I have -- the baseline is the preamp connected to the power line and analyzer, with no power. The 60Hz bump is =140DbV. Input was 400uV and 63dB gain.
An externally hosted image should be here but it was not working when we last tested it.
and a comparison of the different D3a's I have -- the baseline is the preamp connected to the power line and analyzer, with no power. The 60Hz bump is =140DbV. Input was 400uV and 63dB gain.
An externally hosted image should be here but it was not working when we last tested it.
Yes, they were all burned in. Only one tube is problematic as I see it. This pre is in its temporary house -- no umbilical connecting the power supply to the pre so things will improve when that bit of business is being attended to -- in the next couple of days.
My THD% measurements, with the 400Hz HPF are pretty consistently in the 0.03% to 0.04% range and the sound is just great!
Boards came in the other day but I can't solder one up until after Christmas.
My THD% measurements, with the 400Hz HPF are pretty consistently in the 0.03% to 0.04% range and the sound is just great!
Boards came in the other day but I can't solder one up until after Christmas.
Power Supply PCB --
An externally hosted image should be here but it was not working when we last tested it.
