Preamp tubes for cathode followers?

[TheWinterSnow]

So I had band practice today, get set up and fire on my 5150, it plays one quick sound and then goes down in volume as if the post gain was down all the way. Connect up a cable to the FX return and tap on it, sound...slave my other guitarists 5150's pre to my power, it works...I think oh its the send jack, I will clean that, doesn't work...replace the FX loop with his FX loop tube, works. Long story short, the cathode follower slot in the FX loop went to shit. I just replaced my preamp tubes two years ago. Then it dawned on me that I heard about cathode follower slots eating through tubes.

Thing is when I just looked at JJ's datasheet, their ECC83s are listed to have a cathode to filament voltage difference of 180v which is relatively high for more tubes which are only 100v typical. I figured the only way to prevent such failure from happening again is to get a tube that is better built for cathode followers, but 180v is a large number.

So the amp gurus with many years of tube experience, what tubes have you used that have lasted the longest in the cathode follower position? If need be, I am more than willing to step away from 12ax7s for that slot if that means I can get a tube that is more resilient to excessive Uk/f. It probably doesn't help that our drummer place where we practice not only does not have a ground but has constant brownouts does nothing but wear down the filaments. I sound like I am complaining though, but 2 years of going through such abuse when most people can't get JJ's to last a few months is pretty impressive, I just expected them to last longer.

EDIT: From looking at datasheets some more it looks like the 12AU7/ECC82 would be able to dish out the position a lot better. Can anyone confirm this?

 

[Lasse Lammert]

most russians aren't good in CF stages. JJ usually are ok IME though.
Tung Sols can't handle it at all...good tubes, but don't ever put they in the CF stages.
chinese can handle it without a problem.

 

[sprack]

This seems to confirm that the 12AU7 would be a better fit.

http://www.freewebs.com/valvewizard/accf.html

 

[TheWinterSnow]

yup, that seems to sum up what I could gather. Now If I can just get any fist hand accounts of this actually working.

I also just found out by looking at schematics that Peavey decided to daisy chain four stages in series with 23v, which means each tube is getting 5.75V and not the 6.3v they require...sounds like they have purposely attempted to make the tubes die quicker on purpose, that and the CF's heater isn't elevated enough to handle a good portion of near saturation in that position, which probably attributed to the failure. That doesn't surprise me though considering what they did with the SGR's which hinder the life of the power tubes.

 

[lolzgreg]

RUBY 12AX7 HG+ is great in the CF.

Nothing from New Sensor (Tung Sol, Sovtek, EH, etc) will do, because the tubes cannot handle the voltage of the CF stage, as Lasse touched on.

 

[TheWinterSnow]

I'm thinking of going with the JJ ECC802 Gold pin, which is a high performance long plate 12AU7. Anything high performance and brings out more dynamic, fattens up the sound and is not only able to drive the amount of current required in that position but can also handle almost 3 times the dissipation of a standard 12AX7 sounds like a good tube to me.

I can't see myself playing other tubes another than JJ and New Sensor, and sense New Sensor sucks for that position, a high performance JJ should be my next try.

EDIT: Scratch the 802, looks like it can't handle the Vk/f of the CF, the regular 82 looks like the only option.

 

[PaoloJM]

I'm thinking of going with the JJ ECC802 Gold pin, which is a high performance long plate 12AU7. Anything high performance and brings out more dynamic, fattens up the sound and is not only able to drive the amount of current required in that position but can also handle almost 3 times the dissipation of a standard 12AX7 sounds like a good tube to me.

I can't see myself playing other tubes another than JJ and New Sensor, and sense New Sensor sucks for that position, a high performance JJ should be my next try.

EDIT: Scratch the 802, looks like it can't handle the Vk/f of the CF, the regular 82 looks like the only option.

I'm pretty sure the heater to cathode voltage isn't any higher for these valves.
180V is damn high and is more than sufficient for that spot.
The valve probably just went, like any other valve might.

You won't get the the real benefits of using a 12AU7 in that spot unless you modify the circuit itself.

 

[Wolfeman]

Am I the only one who doesn't blow through tubes in a CF spot? I never even knew people had issues until now. My SLO(clone) has had the same set of tung-sols in it for over 4 years now and it sounds A-Ok.

 

[TheWinterSnow]

I'm pretty sure the heater to cathode voltage isn't any higher for these valves.
180V is damn high and is more than sufficient for that spot.
The valve probably just went, like any other valve might.

You won't get the the real benefits of using a 12AU7 in that spot unless you modify the circuit itself.

Yes 180v is as high as they come (however the 802 I mentioned was only 100v). Its not that, but from what I have read from Merlin Blencowe,...

The ECC82 (12AU7) is particularly suited to this job as we do not require significant gain, has good current drive and is capable of maintaining large signal swings into low impedance loads.

and in terms of the circuit, the 5150's setup is very in the ballpark with what would be typical for an AU7, it might not have as much current gain as it could have, but it is in the FX loop which means we just need a low impedance and nothing more and with the load and bias of that slot, it does its job, although biased slightly cold, which actually benefits sensitivity. The following stage to is a very small voltage gain stage so the low mu of an AU7 really isn't hindering anything and again is running slightly colder which again helps with sensitivity.

Am I the only one who doesn't blow through tubes in a CF spot? I never even knew people had issues until now. My SLO(clone) has had the same set of tung-sols in it for over 4 years now and it sounds A-Ok.

You are a lucky bastard for sure. Then again like I have mentioned the power where I play is in a brownout than actually working at the proper voltage. I am also still trying to figure out what voltages Peavey runs their heaters at. Schematics are strange to say the least and it looks like from what I gather that they are supplying 6Vdc per tube, but then there is an unrectified and unfiltered tap connected to the rectifier of the filament tap tied by a resistor and the other part of that tap goes straight to the filaments which is AC. So it looks like AC blended with a filtered DC. I don't know what the filaments are being ran at, but it looks like the CF isn't even elevated as it should.

 

[Wolfeman]

Yeah the PS in the 5150's is a little nuts. They have the AC filament supply, DC filament supply, and the switching supply all mixed into a confusing hodgepodge. I don't know what they were trying to accomplish, it could be done a lot easier.

 

[TheWinterSnow]

Its definately weird, from what I gathered they used the DC filament to raise or offset another AC filament up by 24V so that the CF is an AC tap that is elevated through the DC coupling. Weird, and not elevated enough.

 

[Wolfeman]

I wonder how much quieter the amp would be if it was modded to just run a standard 6.3v AC heater setup to all tubes, with the heaters elevated off the screen supply node like a normal person would do.

 

[TheWinterSnow]

Well the majority of the filaments are DC so it doesn't get any quieter than that, but as you said, taking the elevated from the screen supply would be much more beneficial.