All You Tube Guru's

[R.G.]

Just as an aside, there's not much point in using a tube for a cathode follower any more. The rise of high voltage low(er) capacitance MOSFETs will let you replace a tube as a CF with a MOSFET. The MOSFET Follies article at GEOFEX probably needs updated now, as I was using IRF820s when I wrote it. They work, but some of the later Zetex devices do the same thing in a TO92.

This gives you a triode section to use for gain, which is why we're here.

There are some circumstances where you can make a circuit force a tube follower sound different from a MOSFET follower, but it's not too easy.

[DougH]

Just as an aside, there's not much point in using a tube for a cathode follower any more. The rise of high voltage low(er) capacitance MOSFETs will let you replace a tube as a CF with a MOSFET. The MOSFET Follies article at GEOFEX probably needs updated now, as I was using IRF820s when I wrote it. They work, but some of the later Zetex devices do the same thing in a TO92.

This gives you a triode section to use for gain, which is why we're here.

There are some circumstances where you can make a circuit force a tube follower sound different from a MOSFET follower, but it's not too easy.

Tell that to the guys who swear up and down that tube followers compress and that adds to "the sound". I have no dog in the fight, just sayin'. 

I built an amp using a MOSFET follower for tone stack buffering, to save space, and it sounds really good to me.   

[craigmillard]

Cheers for the input guys, im thinking of giving a mosfet a go on future designs, seems like a good way to free up a triode!

Still trying to get my head around the heaters and elevating them, does the schematic below look right? va is to be connected to the b+ reference of 220k/82k divider to give about 80v from 300v?

Im only raising 1 317... and will run the CF from that?

What do you think?

[MetalGuy]

Just as an aside, there's not much point in using a tube for a cathode follower any more. The rise of high voltage low(er) capacitance MOSFETs will let you replace a tube as a CF with a MOSFET.

Allow me to disagree to some point. If you have an FX type of CF (FX send buffer) yes you can replace it with a MOSFET. However if you have a tone stack type of CF and you use directly a MOSFET it will not behave neither sound as a tube. It requires some additional circuitry to do so:

http://translate.google.com/translate?sl=ru&tl=en&js=n&prev=_t&hl=fr&ie=UTF-8&layout=2&eotf=1&u=http%3A%2F%2Fgtlab.net%2Fgtlab4%2Farchives%2F160

[R.G.]

Of course you may disagree. There are many levels of simulation and emulation of various devices. It all depends on how close you insist on getting to the original.

A MOSFET is not a triode; it has a much higher transconductance, and that in turn introduces into the math what comes out as a lower internal impedance at the cathode - the "Re" from that link. It's actually much closer to a pentode in operation. I suspect that a pentode and a MOSFET would be much more alike than a triode. The diodes add more layers.
[Note - I read the link wrong. The "Re" is the load resistor. The added resistor is the trimmer in series with the source.]

On the other hand, a MOSFET source follower gets you very close. It depends on how much work you want to do.

But yes, you can indeed insert a additional resistor in series with the source to add in a resistance which emulates a triode more closely. It's always possible to 'dirty up' a higher performance part to simulate a lower performance part.

Tell that to the guys who swear up and down that tube followers compress and that adds to "the sound". I have no dog in the fight, just sayin'.  

I have. You can always tell a tube purist - but you can't tell them much if it doesn't match their preconceived notions. Many of them would fight to the death that no solid state device will *ever* sound like a tube. And if you trip them up with a fair test, they'll always believe you are lying about the test.

I built an amp using a MOSFET follower for tone stack buffering, to save space, and it sounds really good to me.    

Me too. I'm sure there are people somewhere that can tell the difference. I can't. Neither could anyone who I asked to do a blind listening test. However, 100% of the people who were told what they were listening to said the tube sounded better.  

[MetalGuy]

This MOSFET TS CF is one of the next things I'll have to A/B for myself.

[R.G.]

Please do. And take it the extra step: have someone other than yourself switch between the two different choices, and record your choices on paper. The trick is to find out two things: (1) do you reliably hear a difference between the two that's more reliable than 50-50 guessing and (2) if so, is it in the correct direction.

The hifi people don't like blind tests after some highly regarded ears tried this and wound up picking the supposedly wrong choice more often than the politically correct one. 

[amptramp]

Has anyone used a regulator as a voltage follower?  Imagine signal going into the adjust pin of an LM317 and the output following faithfully 1.23 volts above that.  A few microamps into the adjust, an amp at the output.  Obviously, the internal feedback will colour the sound somewhat, but wouldn't it be nice to have a cathode follower with current limiting and temperature shutdown that could drive a 50-ohm load?

[StarGeezers]

  Once you put SS diodes in as a rectifier , all the rest of that tube purist  stuff goes right out the window ...       No reason not to try something new... you can always change it if you don't like it ... 

[GFR]

What about this? (effects of a badly biased DC coupled cathode follower)
http://www.freewebs.com/valvewizard1/dccf.htm

Has anyone used a regulator as a voltage follower?

There are many examples of the LM317 as an audio amplifier.

http://www.diystompboxes.com/smfforum/index.php?topic=77872.0
http://electroschematics.com/wp-content/uploads/2012/03/lm317-audio-amplifier.gif
http://www.tubecad.com/2004/image129.gif
http://www.edn.com/contents/images/185953f1.pdf

[craigmillard]

Hey guys,

Any ideas on the above schematic?

Can i just inject the vref for elevated heaters into the ground side of the bridge and get rid of the 100 ohm resistors?

[merlinb]

Can i just inject the vref for elevated heaters into the ground side of the bridge and get rid of the 100 ohm resistors?

Yes.

Im only raising 1 317.

Both regs will be elevated since they're all part of the same circuit. Not sure if that's what you were intending..?

BTW, C5 and C9 are unecessariy large; 10uF-ish is plenty. In fact, you can probably get away with only the ceramic caps on the output, if you're only powering heaters.

[DougH]

But yes, you can indeed insert a additional resistor in series with the source to add in a resistance which emulates a triode more closely. It's always possible to 'dirty up' a higher performance part to simulate a lower performance part.

I built an fx loop recovery amp with an op amp. As expected it really sources a lot of current, and that was making the guitar amp sound a little more "toppy" than usual. So I installed a 100k series resistor and it smoothed right out. Problem solved. Amazing what a 6 cent resistor will do.

[iccaros]

Hey guys,

Any ideas on the above schematic?

Can i just inject the vref for elevated heaters into the ground side of the bridge and get rid of the 100 ohm resistors?

your elevating the regulators, but the heaters do not see the high voltage.
heater elevation works by having a high voltage DC signal ride over a AC current signal. The Heaters will feel both, but the DC part is not giving any real current so it does not add significantly to the heat. But when a positive potental is on both sides it subtracts what the tube sees at those points and lessens arching..

So in my, could be wrong and may just be, elevating the regulator does no do this.

[merlinb]

your elevating the regulators, but the heaters do not see the high voltage.

Think you're getting confused... If the regulators are elevated, and the heaters are powered by the regulators, then the heaters see the high voltage.

[iccaros]

I believe you, just don't see how, as they see the relative voltage right. If I put a meter across the heater does it read the high voltage or the 6v - 12 v

If I elevate my AC heaters with a DC, and I place my meter on the CKT, on DC I will see the higher voltage but at AC I will see the 6.3


So As I said, I could be wrong. But I don't understand why.

[slacker]

With the elevated regs it's basically the same as your AC example. measured from ground, one end of the heaters will be at say 80 volts and the other will be at say 92 volts. So if you measure across the heaters you'll see 12 volts but relative to ground or the cathode voltage or whatever they're at the elevated voltages.

[amptramp]

It is fairly common in tube audio circuitry to elevate the heater voltage because some tubes have the heater at the top and bottom overhanging the cathode and contributing some current to the plate, heavily modified by hum.  To reduce this source of hum, the heater voltage is elevated to about 30 volts above the cathode.

[iccaros]

Thanks Slacker.. That made Sense.. I just was not picturing it in my head..