recommend 12ax7 for 5V heaters?

[martthie_08]

kind of odd question: I need to replace some 12ax7's in an EQ which has a regulated 5V heater supply (measured 4.8V). That is really low, but it is a commercial unit and I won't modify it. Until now I had some Tung Sol in there which sounded fine, but after a couple weeks two of three went out and I need to try something different. So does anyone have a suggestion for a particular type or brand of tubes I fry next?

 

[mjrippe]

Would you like to share what piece of gear this is?  Only reason I can think of for 5v is to keep costs down if the unit has 5v logic or microprocessor circuits as well.  Otherwise look for a trimmer and boost the filament as close to 6.3 as you can get    What voltage does it have on the plate?

 

[martthie_08]

hello, thanks for your reply, I will answer your questions, let me know if there is anything else you'd like to know.

Would you like to share what piece of gear this is?

sure, it is an older Summit 200b EQ, hybrid concept with 990 opamps on the output. It has something special the way it evens out the "digital" sound, works extremely well for softer songs, even without using much of the eq section itself. Not a subtle effect.

Only reason I can think of for 5v is to keep costs down if the unit has 5v logic or microprocessor circuits as well. 

I didn't really dive into the power supply section yet, but I assume they use fixed voltage regulators, I'd have to measure what comes in after rectification, maybe I could put lm7x06 regulators, as I believe this section is just for the heaters. Unfortunately I do not have a precise schematic, even after signing a NDA I only got a crude sketch of the signal path part.

What voltage does it have on the plate?

uhm, I don't recall, B+ was something like 280V I believe

 

[mjrippe]

Ok, so it is not some cheap ART or whatever with a starved plate design.  If it has 990 opamps and a reasonable plate voltage I can't imagine why they would use a regulated 5v supply for filaments.  Of course I also don't see how a LOW filament voltage would cause your tubes to fail.  Did you measure the filament pins for continuity on the bad tubes or did they fail in some other fashion?  Either way, it seems that you should be able to up the filament voltage without problems as long as the transformer tap can handle it.

 

[Gus]

What were the original stack tubes?

 

[ruffrecords]

kind of odd question: I need to replace some 12ax7's in an EQ which has a regulated 5V heater supply (measured 4.8V). That is really low, but it is a commercial unit and I won't modify it. Until now I had some Tung Sol in there which sounded fine, but after a couple weeks two of three went out and I need to try something different. So does anyone have a suggestion for a particular type or brand of tubes I fry next?

I was once testing a 12AX7 in a mu follower and was surprised that I only got a gain of about 50 out of it. I then realised I had wired it for 12V heaters but fed it with 6.3VAC - and it still gave half gain. The  normal tolerance on heater voltage is 10% so they should be happy down to just under 5.7V. 4.8V is nearly 24% low. If my experience is anything to go by they should still work but why the designers should do this I don't know. You do tend to get lower noise with lower heater volts but that's not really a big issues in a line level device.

Are you sure the voltage is 4.8V?  I have seen designs that use a 5V regulator with a couple of diodes in series with the ground leg to get the output volts up to 6.2V dc.

Cheers

Ian

 

[hop.sing]

kind of odd question: I need to replace some 12ax7's in an EQ which has a regulated 5V heater supply (measured 4.8V). That is really low, but it is a commercial unit and I won't modify it. Until now I had some Tung Sol in there which sounded fine, but after a couple weeks two of three went out and I need to try something different. So does anyone have a suggestion for a particular type or brand of tubes I fry next?

I was once testing a 12AX7 in a mu follower and was surprised that I only got a gain of about 50 out of it. I then realised I had wired it for 12V heaters but fed it with 6.3VAC - and it still gave half gain. The  normal tolerance on heater voltage is 10% so they should be happy down to just under 5.7V. 4.8V is nearly 24% low. If my experience is anything to go by they should still work but why the designers should do this I don't know. You do tend to get lower noise with lower heater volts but that's not really a big issues in a line level device.

Are you sure the voltage is 4.8V?  I have seen designs that use a 5V regulator with a couple of diodes in series with the ground leg to get the output volts up to 6.2V dc.

Cheers

Ian

We have a Summit audio micpre in the studio, with a similar topology. Stereo with three 12ax7 at less than 5 volts filament voltage and two 990 opamps. Designed by Dave Hill, so there must be a good reason for the low voltage. I remember reading somewhere, probably in the manual, that it is done for low noise. The original tubes in there where chinese 12ax7, I have similar ones in my guitar amp that say mesa/boogie on them.

 

[pucho812]

I have heard of starving the plate but starving the heater? At that point I would think it's nothing more then a dim light bulb.

 

[ruffrecords]

We have a Summit audio micpre in the studio, with a similar topology. Stereo with three 12ax7 at less than 5 volts filament voltage and two 990 opamps. Designed by Dave Hill, so there must be a good reason for the low voltage. I remember reading somewhere, probably in the manual, that it is done for low noise. The original tubes in there where chinese 12ax7, I have similar ones in my guitar amp that say mesa/boogie on them.

Lower noise is about the only reason I can think of. Shot noise in tubes depends on cathode temperature which in turn depends on heater volts so lowering heater volts will certainly create less shot noise. Might be worth doing in a mic pre but as I said earlier, noise should not be a problem in an equaliser operating at line level.

Do you have the model number and/or specs for the Summit mic pre you mentioned so I can see if it claims an especially good noise spec?

Cheers

Ian

 

[martthie_08]

I remember reading somewhere, probably in the manual, that it is done for low noise.

well, they actually say that higher priced tubes that are regarded superior in other applications performed worse in this circuit.

Are you sure the voltage is 4.8V?

yes, this is what I actually measured

Did you measure the filament pins for continuity on the bad tubes or did they fail in some other fashion?

good call, I will test the tubes in a guitar amp that has the same tubes in it to be sure...

What were the original stack tubes?

I will have to check, we dug them out a couple days ago, but haven't done any testing yet. Got a session running until the end of the week, then I will have more time to get back on this. I do remember though that replacing the stock tubes changed the sound of the EQ in a good way.

thanks everybody!