White Cathode Follower sound

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
thanks that makes sense to me....just thought I would stick my head above the parapet......to see is my understanding was correct. Still got lots to learn but I can follow most of the threads even when it gets technical.
The Potential divider is used to apply dc to the heaters which if I understand correctly are not grounded as they are raised above ground by the series resistance chain. I had thought that the DC supply came into the equation as well.

Doing that I am assuming has an affect on noise and hum as well? In either SRPP or WCF the upper cathode is at anode potential if I get how it works.

Not wishing to side track too much, what is the difference between SRPP or WCF? I know it is open to debate as to whether they may be the same thing.....just wishing for some clarity that is all.

My modus operandi is if you do not ask you will not know but the is rice nearly cooked I do like.
 
The ECC88 was designed as a cascode rf amplifier for VHF receivers so the upper stage cathode is expected to be at least equal to the lower stage plate. This is why its Vhk is 150V. Alternatively you could use the HQ version the 6922 which has a Vhk of 200V for both triodes.

Cheers

Ian
 
thanks that makes sense to me....just thought I would stick my head above the parapet......

Yep, glad you did :)

The Potential divider is used to apply dc to the heaters which if I understand correctly are not grounded as they are raised above ground by the series resistance chain. I had thought that the DC supply came into the equation as well.
Yes the potential divider from the H.T. supply determines how much the L.T. is elevated above ground. What would have been the side of the heaters that was typically connected to a ground reference is now connected to the mid point of your potential divider.
There's no reason you couldn't elevate it by connecting the + side of your heaters (+6V3 or +12V6) instead of the usual ground side, but I don't see any benefit.

Not sure what you mean by DC supply coming into the equation... do you mean the low tension heater supply or the high tension B+?



Doing that I am assuming has an affect on noise and hum as well? In either SRPP or WCF the upper cathode is at anode potential if I get how it works.

If your supplies are clean, then there's no detriment to noise or hum. In some older schemes where AC was used on the heaters, you sometimes saw the heater's mid point being elevated a little by a divider from the H.T. This was to superimpose a little DC on the heater line to make it a dB or so quieter. Supposedly. I don't know the theory or evidence for if/how this worked. I use DC so I never pursued it.

If your heater is elevated such that it's near where the lower tube's anode is DC wise**, (about half of B+ if it's a white follower,), then I don't see it as an issue as long as the lower cathode is within the specified safe range of cathode to heater potential difference.


Not wishing to side track too much, what is the difference between SRPP or WCF? I know it is open to debate as to whether they may be the same thing.....just wishing for some clarity that is all.

My modus operandi is if you do not ask you will not know but the is rice nearly cooked I do like.

Oh boy! Yeah, open to debate as to whether they'r the same, or whether they each have different advatages for different requirements.

I'm gonna go and refresh my brain on a few SRPP details I've forgotten, Ian is the man who knows the ins and outs of those.

** I've never needed to elevate the heaters that high myself.
 
Last edited:
The ECC88 was designed as a cascode rf amplifier for VHF receivers so the upper stage cathode is expected to be at least equal to the lower stage plate. This is why its Vhk is 150V. Alternatively you could use the HQ version the 6922 which has a Vhk of 200V for both triodes.

Ian, I'm assuming you found the new 6922's that you use in your stuff to be within that spec.
Certainly the old stock E88CC's that I'm used to did. But around 10 years ago when I tried what was available new then, good E88CC NOS were getting pricey so..., the specs I encountered (and a few other specs) on the new didn't match what I was used to on the old. Things progress though.
 
Not wishing to side track too much, what is the difference between SRPP or WCF? I know it is open to debate as to whether they may be the same thing.....just wishing for some clarity that is all.
SRPP has gain similar to a common-cath stage and must be optimized for a specific load.
WCF has unity gain, more NFB and thus lower THD.
 
Thanks Abbey,
I'll read Broskie's then, it'll do :)
There was another good analysis/treatment that I'd read but, can't remember who or where now.
 
OK got it in my brain again.

Stevie, the link Abbey gave has the benefits and drawbacks of the SRPP and much more detail than I could write here. I say just read that.

In short, I'd forgotten the details of how you optimize for the specific load in SRPP, nevertheless, my preference for a naked output signal (no output transformer) in a studio environment was/is for the White Follower over the SRPP.

If it's your own studio, you can control, or you know what the load impedances are going to be.
Whereas, out in the world, there is such a discrepancy in what load your output might see - between 600 ohms and 10K ohms - that I'd prefer a beefy White that could handle the low load. It might be wasteful of current sitting idling highly away if it's only driving a 10K load, but it still stays in balance with a 600 ohm load up until the maximum peak current and voltage that it's biased to is exceeded.

If you make that idle current able to handle , at minimum, +20dBu into 600 ohms, then you'll be good all round.
 
Last edited:
SRPP has gain similar to a common-cath stage and must be optimized for a specific load.
WCF has unity gain, more NFB and thus lower THD.
Strictly speaking an SRPP can be optimised for minimum distortion into a specific load but nobody forces you do do this. The change in distortion with load is not huge. I have found in general they produce no more distortion than an SE stage when delivering real power and they do it for half the quiescent current.

Cheers

Ian
 
Ian, I'm assuming you found the new 6922's that you use in your stuff to be within that spec.
Certainly the old stock E88CC's that I'm used to did. But around 10 years ago when I tried what was available new then, good E88CC NOS were getting pricey so..., the specs I encountered (and a few other specs) on the new didn't match what I was used to on the old. Things progress though.
The EH spec says 200V.

https://shop.ehx.com/catalog/addimages/6922eh.pdf

Cheers

Ian
 
I have also used JJ E88CC and I found them to be inconsistent (especially for distortion) and a couple of them failed completely.. I have since standardised on 6922EH.

Cheers

ian
 
SRPP has gain similar to a common-cath stage and must be optimized for a specific load.
WCF has unity gain, more NFB and thus lower THD.
SRPP is similar to Common-Cathode Stages is that another name for a Long-Tailed Pair (LTP)?
I follow what you have all said so far and I am familiar with TubeCad site but hate the way it sends you round in circles......been following the Motown Direct Amp Inspired Preamp topic as well.....I had some ideas as to how it was put together and from what has come out of that discussion the conclusions came as no surprise.

I have my reasons as I have a 5 input mono mixer for which I have no schematic.....I keep putting off the drawing it out by hand.....I do have a use for it.......there is no power supply for it either......I do not think it is much more than a 5 way summing mixer with Preview/cue bus and a couple of gain stages and a monitor amp ECL82 with a Tone shaping equalistion in another outboard box.....a few aux sends and returns as well.....it is cramped and hard to follow plus I need to be in the right frame of mind to draw it out and so far that has not happened.

Is there a reason why you could not or should not use an EF91 or EF94 triode strapped in SRPP or WCF?
Thanks to all those that have helped me with my understanding so far much appreciated...still a long way to go though.......

Can any one recommend a tutorial or how to use spice modelling? By all means direct me to a thread where it has been discussed or is there an easier way to do it? Tube data manuals with standard schematics(circuit blocks) and the data sheet for example? If the formulaes are there then I can figure it out. But theory is a long way from practice.
 
SRPP is similar to Common-Cathode Stages is that another name for a Long-Tailed Pair (LTP)?
Huh?

Is there a reason why you could not or should not use an EF91 or EF94 triode strapped in SRPP or WCF?
Probably not*, but why would you want to do that?

* You need to check if the voltage between heaters and cathode is within specs, though.
 
SRPP is similar to Common-Cathode Stages is that another name for a Long-Tailed Pair (LTP)?

I can see where you're coming from in your thinking (two tubes coupled at their cathodes in LTP) but, when we say 'common cathode' it just means that the cathode is at a common or a reference voltage to the other two electrodes. So it's another way of saying it's a stage where signal will flow out of the anode.

A 'common anode' stage would be another name for a cathode follower.

A 'common grid' has its grid held at some reference and would be used where you have a low impedance input via the cathode which exits the anode. This would be the case with a top tube in a cascode connection. Or taking your long tail pair example - if one of the tubes has its grid grounded then that particular tube is in 'common grid' mode.



Is there a reason why you could not or should not use an EF91 or EF94 triode strapped in SRPP or WCF?
Thanks to all those that have helped me with my understanding so far much appreciated...still a long way to go though.......

Some pentodes make quite nice triodes and are still quite inexpensive for NOS pieces. I haven't looked at EF91 or EF94 but sometimes the data sheets give triode curves.

However, as I mentioned above, I think high Gm pentodes kept in pentode mode would be killer in White Followers.


Can any one recommend a tutorial or how to use spice modelling? By all means direct me to a thread where it has been discussed or is there an easier way to do it? Tube data manuals with standard schematics(circuit blocks) and the data sheet for example? If the formulaes are there then I can figure it out. But theory is a long way from practice.

I use spice modeling but, actually, not for tube stuff.
Ian does, as probably has Abbey, and a lot of others.
They'd have more info re. using it for tube modeling.
 
Last edited:
I have also used JJ E88CC and I found them to be inconsistent (especially for distortion)

Yes, thanks. I'm about 90% certain it would have been JJ as I had some of their E99CC's around too.

Distortion was the main issue I had too.
Mystery solved, EH it is then.
 
Huh?


Probably not*, but why would you want to do that?

* You need to check if the voltage between heaters and cathode is within specs, though.
Because I had not thought it through and forgot that you can use Pentodes in WCF. Thus my conclusion was that the pentodes should be operated in Triode Mode.
 
I can see where you're coming from in your thinking (two tubes coupled at their cathodes in LTP) but, when we say 'common cathode' it just means that the cathode is at a common or a reference voltage to the other two electrodes. So it's another way of saying it's a stage where signal will flow out of the anode.

A 'common anode' stage would be another name for a cathode follower.

A 'common grid' has its grid held at some reference and would be used where you have a low impedance input via the cathode which exits the anode. This would be the case with a top tube in a cascode connection. Or taking your long tail pair example - if one of the tubes has its grid grounded then that particular tube is in 'common grid' mode.





Some pentodes make quite nice triodes and are still quite inexpensive for NOS pieces. I haven't looked at EF91 or EF94 but sometimes the data sheets give triode curves.

However, as I mentioned above, I think high Gm pentodes kept in pentode mode would be killer in White Followers.




I use spice modeling but, actually, not for tube stuff.
Ian does, as probably has Abbey, and a lot of others.
They'd have more info re. using it for tube modeling.
Thanks that all makes sense I was checking that you were not using another name for the same type of circuit. People often use many names for the same thing not always the common name.
 
The heater to cathode voltage rating varies widely among tube types - and is why Jensen's AS021 (see https://www.jensen-transformers.com/wp-content/uploads/2014/08/as021.pdf) application schematic uses a 12BH7 in its "totem-pole" output driver. From the "RCA Receiving Tube Manual, RC-20, 1960" data sheet for 12BH7:
Peak Heater-Cathode Voltage:
Heater negative with respect to cathode ... 200 max volts
Heater positive with respect to cathode ... 200 max volts (the DC component must not exceed 100 volts)
 
Back
Top