WD Phono 3 - DC on Output?

[Phil Y]

Hello moley,
I am not quite sure what you mean by "before and after" the resistor. A load resistor is connected between +ve signal out and ground.

My thought was that with very little load, the low frequency cut off of the o/p cap will be very low frequency and you therefore might just be seeing H.T. drift being coupled to the output. If so, it would disappear with a "real world" load as the cut off point would be raised above it.

Just a thought and I am more than happy to be told that this is total horse feathers

As Richard says, it would be useful to know roughly what the frequency is.

Phil.

[moley]

Thanks again for the replies.

I plugged in a couple of shorting RCA plugs into the input of the Phono 3 and removed the tonearm cable.

For a dummy load, I put a 50k resistor across the hot pin and ground on the interconnect that is plugged into the output of the phono 3.

Measuring the DC voltage between the hot RCA pin (positive test probe placed at RCA pin prior to the resistor) and the chassis earth I got A similar flickering reading which changes probably 2-3 times a second between roughly -100 and 100mV sometimes peaking a little higher either way. This is a little lower than when I test without a resistor dummy load (approx -250mv to 250mv).

Measuring with the positive test probe placed after the resistor close to the RCA ground gives a steady result of 0.02mV. Is this basically the reading that counts, i.e. under load the DC drops to such a low level it isn't a problem?

[Richard]

I plugged in a couple of shorting RCA plugs into the input of the Phono 3 and removed the tonearm cable.

For a dummy load, I put a 50k resistor across the hot pin and ground on the interconnect that is plugged into the output of the phono 3.

Measuring the DC voltage between the hot RCA pin (positive test probe placed at RCA pin prior to the resistor) and the chassis earth I got A similar flickering reading which changes probably 2-3 times a second between roughly -100 and 100mV sometimes peaking a little higher either way. This is a little lower than when I test without a resistor dummy load (approx -250mv to 250mv).

Yes, this is the way to load it and the correct position for the probes.

Measuring with the positive test probe placed after the resistor close to the RCA ground gives a steady result of 0.02mV. Is this basically the reading that counts, i.e. under load the DC drops to such a low level it isn't a problem?

No, this puts both probes on ground so you would expect zero.

What meter are you using? Digital or analogue (needle)? Is it auto sensing type or have you manually selected the range? Is the range set to DC mV or AC mV? If you are simply holding the load resistor in place consider that your body and movement of the connection may cause bad readings as the meter is very sensitive in this range.

[Richard]

Meters are very sensitive in this range.

I have just tried pressing the probes of 2 digital meters set to DC and AC mV ranges onto my thumbs and I get a similar result to you Moley.

The readings change about 3 times per second flicking plus or minus to over 100 mV depending on movement, how moist my thumbs, are and how hard I press them onto the probes.

Try your meter on your thumbs and see if it is this that is giving the spurious reading. If so arrange soldered or clamped connections and test again with the meter isolated from your body. Also try another meter.

[moley]

Thanks again for the repy.

It is a Maplin digital Multimeter and is set to DC, the range is set to its lowest which runs up to 200mv.

The test probes were clipped onto the test points via aligator clipped test leads so I was not holding these.

I have another multimeter somewhere that I can try, I could also try more sturdy / anchored test points.

When I conducted this test on my Beresford Caiman DAC I get a steady result of under 0.2mV, I did not test with a dummy load however.

I have attached a couple of photos of test leads / clips and the multimeter.

[Richard]

Well that set up looks fine and you've used it on the dac so your method seems good.

Do you have a swinging needle meter, or a scope, which may show how the voltage is changing?

If the output coupling caps are good they should only be passing AC but I'm not sure the voltage would have to swing through 0V (-/+ wrt ground) to be passed.

Before that,

have you swapped the ECC83s? A noisy valve or socket may be a cause, if you don't have any spares swap the last cathode follower valve with one of the others to see if the results change,

what power amp are you using and can you see any LF movement of your speaker cones? (I'm interested that you don't hear the noise but if you have a valve power amp it may be too low f for the output transformers to couple it to the speakers.)

Then,

do you have any 250V (or greater) 0.47uF or 0.22uF plastic film caps? The 2.2uF give a 1.4Hz coupling into 50K. They would move it up to 7 or 14 Hz to see if a voltage is still present.

[moley]

Hi Richard, thanks agaon for your help, there is a bit go off here and take away to try on the test bench.

I don't currently have access to a scope or an analogue needle swing meter.

My Valve Amp is an Almarro A318B SET amp (2x6C33C, 1x6SL7, 1x6SN7). Last year i swapped out the Alps 250Kohm Pot and fitted a stepped 100Kohm shunt pot. The input impedance must be around 100kohm meaning that the corner frequency with the 2.2UF in there at the moment would be 0.7Hz if I have used the high pass filter calculator correctly.

I think I do recall seeing a little cone flap at times but generally leave the grills on so have not notice this or really been looking out for it either. Certainly this something I will have a look at, maybe with some off center pressings.

When I first built the Phono 3 I had intended to revisit the subsonic filter options but have long since forgotten all about that. This is probably a good time to look again and hopefully kill two birds with one stone. It certainly seems like the filter is set too low for my application. Surely anything under 20Hz from vinyl is just noise?

One thing that occurs to me is, would it be possible, prior to removing the 2.2UF Caps, to simulate a change in the filter by simply lowering the dummy load which will give a higher crossover frequency? I could then measure the DC again and see if it is still there.

I might have some smaller poly caps in my box of parts I will have to have a dig around. Certainly it will be a shame to have to remove the MR Clarity Caps but perhaps needs must.

I don't have spare valves but I can certainly try swapping them around first to see what happens. They are TJ Full Music ECC83's which were sourced via ebay from China. I guess it is very possible that the quality of one or more could be suspect.

Thanks again.

[Richard]

Yes, the cap and load values work oppositely but in direct proportion to the frequency. So 5k load with the 2.2uF caps will give 14Hz corner f.

This is fine to test but bear in mind these filters are cumulative through the phono/pre/power amp chain and it would be easy to end up with light sounding bass.

Phono 3 uses large output caps to make a low frequency coupling into most common loads (typically 1.4Hz into 47k). Phono 3S then offers a switchable LF roll off at the first stage coupling. I've built and used both and found it sounded bass light with the filters switched in. Like you though I was using a valve power amp which would have been rolling off the lowest bass anyway.

It would be good to find out the cause of this voltage. Peter Comeau launched the 3S version and also included additional caps and resistors in response to some reported instabilities. I never built Phono without those extra parts so don't know for sure but feel you might want to try adding them.

[moley]

Thanks for the extra information. Yes I only just realised that the subsonic filter was at a different Cap position on the circuit. I am struggling to find a schematic on the 3S so that I can find out what the variations are aside from the subsonic filter switches.
Do you have a link handy? Thanks!

[Richard]

No probs, I've emailed it so you can compare.