Paul Ruby mod in Vox Night Train

[frus]

Back up and read the whole thread.

I'm referring to this:

Also, the PR mod clips the negative swing of each PI output. This MUST happen when the power tubes are driven into cutoff or the diode clipping will be heard. Using a PPIMV will cause that to happen.

If I put the diodes "after" the MV pot (wipers to ground), then it seems to me that they do what they are supposed to (according to Paul Ruby mod), regardless of PPIMV, because they clip the signal after the MV pot, and only when the power tubes are in cutoff. Unless I'm missing something?

[Chuck H]

Hah... I've mentioned before that we need a "slap forehead" emoticon.

I thought you were saying that the PR mod WON"T work with a PPIMV. Which we had already discussed. And I'm so use to looking at my own diagrams and schems, where the PR mod zeners are piggyback on the bias resistors, that I hadn't considered placing the PR mod post master volume. Of course that should work. I would say it's clever, but actually I think I was just momentarily dim.

Are you getting this Silvio?

As to the cathode zener... At 318 Vp you can beat the piss out of el84's. It's not uncommon for cathode biased amps with a Vp in the low 3XX to idle them just over 14 watts! I'm sure you could use a 13 volt zener without smoke. A 13 volt zener is what I use in my design, which has 355Vp.

As to how much more current the psuedo fixed bias thing is causing... Dunno. I never bothered to check. But since the amp is operating fixed bias when the zener is conducting you could check current the same as you would with any fixed bias amp. However, you obviously can't check idle current when the tubes aren't idling!?! But it's safe, trust me.

One thing I never have been able to wrap my head around is how the cathode voltage rise is cooling the bias. I know that the more negative the grid is in relation to the cathode, the cooler the bias. But I also know that if the voltage rises on the cathode resistor that there is more current!?! So let's speculate a typical 2Xel84 cathode biased amp where 340Vp drops to around 310Vp with the amp cranking. And the 10V on a 130 ohm cathode resistor rises to about 20V. My pea mind says that there is 20V across 130 ohms, so 154mA. And 154mA X 310Vp = 48 watts. Divide by two and you get 24 watts of dissapation per tube! Obviously there's something I'm not getting because these are really typical numbers and no el84 can dissapate 24 watts. Hopefully someone will see this and chime in to explain and we'll both have an opportunity to learn.

[Silvio55]

Ah! So I just put the diodes across the pot´s wipers and ground! Great there frus! I hope this help to get rid of the mosquitos!!

[frus]

there is 20V across 130 ohms, so 154mA. And 154mA X 310Vp = 48 watts. Divide by two and you get 24 watts of dissapation per tube! Obviously there's something I'm not getting because these are really typical numbers and no el84 can dissapate 24 watts.

The tubes generate 48W of power, but it doesn't all go to dissipation (heat = waste), at that time it's dishing out maybe 20 watts to the speaker, so only 28W gets dissipated

[frus]

Ah! So I just put the diodes across the pot´s wipers and ground! Great there frus! I hope this help to get rid of the mosquitos!!

It surely helped my 18W
Good luck!

[trobbins]

The cathode bias resistor sets the idle current as a function of the plate-cathode voltage-current loadline - that loadline has a 'negative resistance' type characteristic (ie. lower anode current for higher anode-cathode voltage and higher control-grid voltage) - similar/same as doing a loadline for a preamp stage and showing the cathode resistor lines crossing the valve loadline at the operating point.

With a capacitor bypassed cathode bias, then as the bias voltage is pumped up by the large signal swing in plate current - if you suddenly stopped the signal, the idle current would be lower, as the bias voltage is higher, and the cap would discharge.

If you have a constant 20V of bias across the 130R, due to the cap, then you do indeed have redplating going on!

Edit:
That last sentance is not correct. If for example the valves were saturating and the signal was a square wave, then the valves would be dissipating the current of 154mA, but their voltage would be the saturation voltage. Most of the 48W would be transferred to the load.

[Steve Conner]

Yes, when applying the Paul Ruby mod to a PPIMV, the logical place for it is right on the power tube grid after the master volume.

Chuck, the cathode zener doesn't make the bias hotter, it just reduces the tendency for the circuit to bias itself colder under overdrive. It's not shovelling fuel into the furnace so much as closing the windows.

Another way of looking at it is: even though the voltage across the cathode resistor is less than the non-Zener case, the total cathode current could be greater, once you add in the current flowing in the Zener.

[Chuck H]

Chuck, the cathode zener doesn't make the bias hotter, it just reduces the tendency for the circuit to bias itself colder under overdrive. It's not shovelling fuel into the furnace so much as closing the windows.

Thanks. When I said "extra current" I only meant that amount that results from the bias not shifting colder. I do get some of this. But as far as the load getting the power that tube is not dissapating??? I guess I always viewed it like a see saw with the tube on one end and the speaker on the other. Is this not the case? So, the power being made by the tube is not directly reflected by the speaker?

[J M Fahey]

Hola Silvio: ELKO (Constitución 3040) tiene todos los Zener que se te ocurran
La página de los de 12/13/15/16V en 1/2W (no necesitás más disipación) es:
Elko Arrow Componentes Electrnicos (Elko)
*Si*necesitás de 5W para algo:
Elko Arrow Componentes Electrnicos (Elko)

NOTE: no need to translate, just a suggestion of a local supplier for Zeners.

[J M Fahey]

...

[Bruce / Mission Amps]

Hola Silvio: ELKO (Constitución 3040) tiene todos los Zener que se te ocurran
La página de los de 12/13/15/16V en 1/2W (no necesitás más disipación) es:
Elko Arrow Componentes Electrnicos (Elko)
*Si*necesitás de 5W para algo:
Elko Arrow Componentes Electrnicos (Elko)

NOTE: no need to translate, just a suggestion of a local supplier for Zeners.

J, is that 1/2w dissipation with respect to the zener at the grid, not across the cathode biasing resistor?

[J M Fahey]

Yes. Only grid Zeners.
Same as you, I was worried that it might be otherwise, but the Mod schematic I found only shows 9.1V Zeners in series with 1N4007.
Don't know whether there are different versions all called Paul Ruby Mod.
In fact, I'd suggest the OP (here and in any other thread) to post first and foremost the schematic he's talking about, simply to save time and to make sure we all talk about the same.

[Chuck H]

The PR mod is not specific values, but a method. The zeners are valued to conduct when the power tubes are in cutoff giving the grid circuit more time to discharge and prevent excessive crossover distortion due to grid loading. So the zener value would be chosen based on the bias voltage, a volt or two higher. Now, since a cathode biased amp increases bias with current I figured that this should also require an increase in the PR zener voltage. But someone corrected me on that once and now I can't rember the details. Just the same, the zener voltage is dependant on when the power tubes are in cutoff so that no diode clipping will make it'self evident in the final sound. Since I prefere fixed bias for clipped tones anyway I use a zener parallel to the cathode resistor with that zener voltage set to stop the voltage rise just past the onset of clipping. Then I choose the PR mod zener value a volt above the cathode zener value. It seems to work.

BTW, here is the Vox Night Train schem. Prowess indicates it is updated. I understand there are mistakes on the original. I can't vouch for this schem though since I've never seen the inside of one of these amps.
http://www.prowessamplifiers.com/sch...n_Updated.html

[Silvio55]

Juan Manuel gracias por la data!! Cualquier cosa que necesite te chiflo y me avisas donde lo puedo conseguir jajaja, aca en mi pueblo se me complica bastante con ciertas cosas!!!
Ok about the schematic, the updated prowess schem is wrong, the values are all right but some components are mislabeled.
I did the mod today, put the diodes (a 1/2 watt 12 volt zener and a 1N4007) across the wipers and ground of the volume pot and it worked!! no more nasty crossover distortion sound!!
I´m really happy now, that problem was bothering me since I bought the amp, now I can really cranck it!!!
Thank you all guys!!!

[Silvio55]

Ok I did further testing and maybe the voltage of the zeners is too low, I can hear the clipping of the diodes, now I have to get some 13 and 15 volts zeners and try again...