The Phonoclone and VSPS PCB Help Desk

[rjm]

To prove it could be done I put 1000uF of extra filter capacitance on each rail and re-measured the output noise.

There you are: the OP27 noise, and nothing but, up to 10 kHz. (Its looks clean up to 100k, as a matter of fact, but my measurement system has some noise spikes up there so it doesn't present well.)

And on the seventh day... zzzzzzzzzzzz.

/R

[GregD]

Originally posted by rjm
To prove it could be done I put 1000uF of extra filter capacitance on each rail and re-measured the output noise.

Am I correct in assuming that this is on the power supply rails before the regulators? Just to make sure.

[rjm]

Yup. 1000uF after the diodes, 100uF in front of each Vreg, 100uF after. All are Chemi-con KMG standard type electrolytics.

[Panelhead]

There is a pair of 4 pole Jensen caps inbetween the diodes and the regs. They measure 33,000 ufd each. There is a 100 ufd low impedance cap on the input and output of each regulator also.


George

[rjm]

Although adding the extra 1000uF of capacitance improved the measured noise performance, subjectively it completely killed the sound.

I dont know why. Have to do more tests first. The difference however was quite striking. The musical presentation was more controlled, solid, with better resolution of low level detail - but by the same token caught a serious case of "hi-end blah". Sounds good - but really, really boring. Like someone shaved all the sharp edges off. Lost all sense of rhythm.

So, like the gainclone, I have to recommend no additional capacitance in the power supply be included. How much filter capacitance (before the regs) I havent worked out yet. To my ears 100uF per regulator is fine, but up to 470uF would help to reduce the ripple harmonics at the output.



/R

[Panelhead]

I replaced the 100 ufd ultra low impedance caps after the regs with 47 ufd Silmics. This got everything working properly.
The gain needs dialing in now. It is a little too high. Used a 1.1K resistor with a 14 ohm cartridge. It works fine, but may benefit a little by going with a 700 ohm of so to drop the gain on the first opamp.
It is very sensitive to input lead dressing. I have been playing with this for a while now. Hum and RFI come and go. Moving my shielded phono cables to get away from the input power cable solves it. Both are shielded, so this is a mystery.
Only got about 5 lps played so far. It is fast and clear. Very good dynamics.
This one has a bunch of capacitance before the regs. It maybe the 4 pole caps react differently than a 2 pole with chips. Distance may be a factor. The big caps are in with the transformer and diodes in another box 3 feet away.
The only weakness so far is a lack of warmth. This may be a new parts problem, hum or RFI issue, or just comparison to my usual phono stage. The Pearl with headamp is overly warm and ripe.
But very good results even with minimal burn in. Expect it to get even better with more playing.
Good job Richard!

George

[rjm]

I still think you should remove those 33,000uF caps, though...

[Franz G]

Richard

I just built a second phono stage, designed by you: the VSPS, to play a little bit with different input-Z in conjuction with my DL103.

I will compare it with my phonoclone. It has a socket for the opamps to compare different opamps.

As I will be a member of the next European Triode Festival in beginning of december, it would be a nice idea to prepare a VSPS for the phono shootout. Imagine a IC-amp competing to a triode pre for example made by Steve Bench...

Could you please simulate or suggest a configuration for the Benz Ebony HO cartridge?

Here the technical part of the announcement from the shootout:

the phono stages must comply with the following specs:

1. gain has to 40db or more (the follwing line stage will have an attenuator at the input so that all phono stages will be listened to at the same volume referenced to 0db@1kHz

2. the cartridge used for the shoot-out will be a Benz Ebony HO, known as "the best high-output moving coil of the world". it has an output voltage of 2.5 mV into 47k (3.45 cm/sec with JVC TRS07). the turntable will be an EMT 938 with a Fidelity Research FR64s tonearm.

3. the phono stages must accept a male 5 pin DIN connector for the incoming signal (pin 1=r+, 2=r-, 3=GND, 4=l+, 5=l-; numbering clockwise viewed from the soldering side). we will provide an adapter for RCA-DIN, but you might not want to compromise the sound of your phono stage with this adapter... important: the wiring of the tonearm is balanced and ground-free - pls take this into consideration when you build the phono stage and--very important!--you *must* provide a ground on your phono stage (banana or spade lug)

4. the phono stages must run either on 230 vac/50hz or on batteries. we will not provide a line conditioner, step-up x.former for 115vac or similar - if your phono stage needs clean juice, make sure the power supply filters out mud and crud.

5. the output of the phono stages must be available on either RCA, CAMAC or XLR, signal can be balanced or unbalanced

6. there will be an attenuator unit between the phono stages and the following line stage. this attenuator with z in=10k ohms i.e. the phono stages must be able to drive this load. the input of the line stage is rc-coupled. (subject to change, but the 10k more or less stands.)

7. if your phono stage hums - tough luck...

Kind regards
Franz

[rjm]

Imagine a IC-amp competing to a triode pre for example made by Steve Bench...

OUch!

If you are dead set on doing this, however, there are no special modications required. The standard VSPS, 47k input resistance, 40dB gain, and the usual ground connections, will suffice.

As you know, the circuit lives or dies on the power supply... with some additional tuning possible though parts selection in the circuit itself.

I have plenty of ideas ... feel free to email me and discuss.

Richard

[Panelhead]

Finally tried lowering the gain of the first stage to 45 or so. This is using a 618 ohm resistor and the listed 14 ohm impedance of the Denon DL-103R.
Used a 150 pf cap across pins 2 and 3. This was discussed early in this thread. There is a desolate area to my west with all the Houston TV transmission towers. It is only maybe 8 miles as the buzzard flies.
Never had this problem with a couple of transformer boosted tube phonos and the current all solid state unit. I think I know why.
My knowledge of opamps is minimal. So it may be a trick coming from using the inverting input.
But looking at this circuit it looks to have infinate impedance. Not zero. What is zero is the loading. The high input impedance sets you up for hum and rfi issues.
Soundwise, this is very good. Still have a little hum to chase. But very live and dynamic presentation.
Walking around the room causes the level of hum to go up and down. This should be fixable. The unit appears to be very quiet on its own.
Now to get another 20 hours or so of playing on it.

George