CCS rules supreme!

pre65 · Post by **pre65** » Fri Nov 23, 2007 8:51 am

Hi-right,with your details,and looking at the datasheet for a current regulator i have made a diagram of how i interpret it.

This is also for that person (whoever he is-there must be one) who knows less than me but would like to have a go.

Andrew · Post by **Andrew** » Fri Nov 23, 2007 10:38 am

Looks about right to me Philip...

Cressy Snr · Post by **Cressy Snr** » Fri Nov 23, 2007 11:47 am

Nick wrote: Steve: At 70v I would have thought you were in the range of a pentode CCS?

Hi Nick
Forgot about how good pentodes are at CCS duties. Should be no problem implementing a pair of pentodes as CCSs on the output valves. A couple of EL34s will handle the current very nicely and at zero additional cost, as I already have the bits.

Only fly in the ointment IIRC would be the need for a negative HT supply for the sink.

Better get reading up on their implementation.
Steve

andrew Ivimey · Post by **andrew Ivimey** » Fri Nov 23, 2007 1:14 pm

The pentode was where I started to get interested in CCS.

The irony of using a pentode as CCS in the Rankin DC CC amp was splendid as this particular pentode could easily be an attractive and more powerful output valve than the 45 which had enslaved it!

hm,

Post by **Nick** » Fri Nov 23, 2007 1:25 pm

Only fly in the ointment IIRC would be the need for a negative HT supply for the sink.

Maybe a couple of pp3's to bias the grid?

Cressy Snr · Post by **Cressy Snr** » Sat Nov 24, 2007 12:35 am

Hi Nick,

That's one way around it.

One of the many good things I found when using the LM317s as CCSs in the input and driver long-tail pairs has been the fact that the perfect balance obtained has killed most of the power supply noise stone dead.

Once the power stage is sorted I'm fairly confident that the very low level remaining noise should be banished. Also looking at the characteristics of the 6as7s I'm using, the horizontal load line resulting from LTP operation with a CCS should make a significant improvement to the linearity of the output stage.

Nice

pre65 · Post by **pre65** » Sat Nov 24, 2007 11:09 am

Hi-this morning i have fired up 832 to take some measurements of cathode voltage.The cathode connection is common for both halves of the valve.

The voltage has stabilised at a steady 20.3 into a 270R 6W resistor so seems to be drawing a STEADY 75ma.

Now,if this current draw is steady,do i need a constant current source,cos surely thats what i have got ?

Feel free to explain how i've got it wrong !!

EDIT.i have seen the voltage vary giving currents from 71.8ma to 74.8ma.What variation would normally be seen ?

Would using a non DMM give me a better idea of the voltage (and therefore current) fluctuations ?

Andrew · Post by **Andrew** » Sat Nov 24, 2007 11:14 am

Hi Philip,

The cathode voltage should go up and down when you put signal in, try it with music on the input.

cheers,

-- Andrew

Post by **Nick** » Sat Nov 24, 2007 11:23 am

The CCS will work like a much larger cathode resistor, so the variations of voltage will be much less. But the average voltage you measure will stay the same, you would need to use a scope to see how it varies in use.

pre65 · Post by **pre65** » Sat Nov 24, 2007 11:25 am

Hi-initially i measured without music but when i started playing the voltage hardly changed.

I have found an old Maplin multimeter with a conventional dial and the voltage swing is the same with,or without music playing.

Still within the range 72.5-74.8ma.

pre65 · Post by **pre65** » Sat Nov 24, 2007 11:38 am

andrew Ivimey wrote:I suspect the theory is the same Nick. Those amps are a bit posh looking though.

Philip, it would be easy to implement on the 832 ... up to a point. First measure the voltage cathode to earth of the output valves. The LM317 will have a breakdown if the difference is more than 40 volts across the input /output. It should be safe but I haven't measured it. Then you have to decide on how much current makes these valves sing, I don't know. Then, there is 1.25volts difference between pin 3 and output, so 1.25 over R1 (what you have to decide on) equals the current in milliamps. So e.g if you want 80ma the resistor you need is 15ohms. I used 18ohms in the el84 amp you heard for 69ma - which I am told is nice and safe for el84s.The CCS replaces anything you have on the cathode to earth (cap and resistor) The cathodes are effectively tied together and go down to earth through the CCS. Now... thinks are there two separate cathodes? If not, they are already tied together, just do the same and listen to teh result. I would guess that by varying the current, using differnt resistors, you will vary the output some and tone, quite a bit.

-worth a try! I'll get round to it but I am rather busy at the mo'.

Can i assume that as the valve is drawing 75ma (average) with a cathode resistor that is what i should be aiming for with a ccs ?

If so that gives me a CCS resistor of 16.5R (2 X 33R would do).

Cressy Snr · Post by **Cressy Snr** » Sat Nov 24, 2007 12:22 pm

In a word Philip.......yes

Or..... you could peruse the characteristics for the 834, look for a current draw that will give the max linearity and use that by recalculating the resistor value.

Don't get hung up over voltages. (Nick cured me of that one a few months ago) Bias current is the important parameter. Your output transformer primary and HT set the anode voltage. The CCS sets the current going through the valve and clamps it firmly at that value regardless of what the music signal tries to do with it

What you end up with is a pure voltage fluctuation on the positive and negative halves of the music signal. The current in both phases is forced constant so there is no possibility of the currents going out of balance. This in turn ensures that there is no DC unbalance in the output transformers and you get perfect second harmonic distortion cancellation and the lowest possible third present in the output.

The third can be reduced to its lowest possible level by judicious choice of current sensing resistor and detailed perusal of the valve characteristic curves, plus a pair of ears.

Believe me, it is worth it and the differences are not subtle. Of course Nick and Andrew up North have been into this sort of thing for years now. A lot of their advocacy of solid state support circuitry has, in some quarters, been falling on deaf ears up to now. Not any more

Steve

pre65 · Post by **pre65** » Wed Nov 28, 2007 12:06 pm

Hi-i am looking at the specifications for the 832A power tube to find out (or try to) the specified maximum cathode current and see how it relates to the 75ma that it is passing with the cathode resistor in place.

I'm having trouble with the terminology,would DC plate current be what i am looking for ?

andrew Ivimey · Post by **andrew Ivimey** » Wed Nov 28, 2007 12:11 pm

In a word, Philip, Yes! what goes through the anode comes up from the cathode, so to speak, so yeah go with that.

Be cautious if you wish and go for 60ma or so - shouldn't make any difference and when you have done it you have saved me so I can get on with 'other things'!

remember, I have found you can go too far with this CCS malarkey and, IMHO, your 832 amp sounds just dandy as it is.

pre65 · Post by **pre65** » Wed Nov 28, 2007 12:25 pm

Thanks Andrew-i think i will go for 75ma as i now have a resistor combination to give me the 16.66R i was looking for,and 75ma is what it seems to draw at the moment.

I agree that it sounds bloody good but it will be easy to reinstate the current arrangement.(scuse the pun)

audio-talk

CCS rules supreme!

#16

#17

#18

#19

#20

#21

#22

#23

#24

#25

#26

#27

#28

#29

#30