12AX7's like the tung sol RI that can't handle cathode follower voltage

[12AX7]

The question i have is about several 12AX7;s that are not recommended for used in the CF position or they will fail. Had that happen in the past at least with the tung sols. Heres the question....since that happened to me i have DC elevated the heaters by i think 30v if i recall and what i want to know is will that allow me to safely used them in the CF position?

[Willabe]

Might not be enough, depends on how high the dcv on the K is.

[12AX7]

Is "k" cathode? It's 160vdc.

[John]

Is "k" cathode? It's 160vdc.

Yes. And if you're elevating the heaters, might as well make it 60-70 vdc. Give you a nice margin. IIRC, a 470k over 100k will get you somewhere in that ballpark with "normal" B+ voltages. But do the math in case I'm wrong.

[HotBluePlates]

The question i have is about several 12AX7;s that are not recommended for used in the CF position or they will fail. Had that happen in the past at least with the tung sols. ...

Got a link to the recommendation/warning?

Electro-Harmonix has the Reflektor factory make the "Tung Sol" tubes, as well as many of the other EH/Sovtek line of tubes.  EH's Tung Sol website includes data on the 12AX7 page that is really just a scan of the original Tung Sol 12AX7 data sheet.

And that original data sheet claims 180v as the max allowable heater-to-cathode voltage.

If you had failures with a new Tung Sol-ski at less than 180v heater-to-cathode, it would probably be an issue to take up with the retailer/distributor where you bought them.  The failure would likely have been due to tube defect or insufficient testing, and the same could happen for any brand of tube.

[12AX7]

I have no idea what the voltage was when the tung sol failed because i didn't have elevated heaters then.

[HotBluePlates]

You mentioned it 160vdc now, so probably wasn't much different any other time unless you've radically changed circuit values in the cathode follower stage.

[jjasilli]

Ditto.  V_k = 160; V_heater = 6.3 x 1.4 = 8.82V_peak.  (Note that for heater :: cathode, DC is used or AC peak).  160 - 8.2 = 151.8V which is within 180V spec for tung sol.  However typical 12ax7 (see GE chart) says 100V spec.  So a 60V elevation might be the safer road, per prior suggestions. I.e., 68.2V_heater (with elevation) is within 100 volts of 160V_k.

[HotBluePlates]

...  However typical 12ax7 (see GE chart) says 100V spec.  ...

We're getting into a gray area of "what sheet from what date?"

G.E. 12AX7 says 180v.
G.E. 12AX7A says 200v.
Sylvania 12AX7 says 200v.
Tung Sol 12AX7 says 180v.
Brimar 12AX7 says 250v.
RCA 12AX7A says 200v.

Philips 12AX7S says 100v, but this may be artificially-low to meet claimed low heater-to-cathode leakage current.

The best rule of thumb may be to split the difference between cathode voltage of normal grounded-cathode stages and the cathode voltage of follower stages.

[jjasilli]

Yes. Re the highlighted GE chart: with heater negative to cathode, there's a 200V spread.  I guess that's the condition we have here, because the CF heater is "less positive" than the cathode.  OTOH Tung Sol spec -- 180V --  does not differ if heater is positive vs. negative to cathode.  But if we allow 20% variation for an actual tube from Tung Sol bogey spec, then unelevated heaters at 8.2V_peak are now out of spec.  So, 30V elevation is better than -0-.  60V better yet.  (60V elevation seems to be the typical max for 12ax7's, probably due to current draw considerations.)

[Willabe]

(60V elevation seems to be the typical max for 12ax7's, probably due to current draw considerations.)

I've been going with 70dcv to 80dcv, in all my builds and on my bread board, as per TUT books, Kevin O'Connor.

I've had no problems so far. 

I do keep the resistor to ground in the voltage divider at 100K for the least loss of mA's. There's a spec on only some sheets that list 100K max. (I think KOC and Merlin talk about it in there books/web site?)

[jjasilli]

Merlin's website touches on the topic of heater elevation, going so far as to say it's required for the CF; but lacking details.  Merlin's Preamp book goes into the CF in great detail with much theory and variations, but no mention of heater elevation that I can find!

[Willabe]

From Merlins web site, heaters section;

"The elevation voltage can be taken from a potential divider across the HT (it doesn't matter where you position the divider), and an elevation voltage around 30 to 60V is typical. The divider should have a fairly high resistance so as not to waste current, although the lower arm (R2) should not be excessively large or Rhk(max) may be grossly exceeded, so it is advisable not to make it greater than 100k. The elevation voltage should be decoupled/smoothed with an arbitrarily large capacitor (C1), say 10uF or more."

I seem to remember someone saying that not all the 12A _ 7 spec sheets have Rhk listed on them, but the few that do list 100K as max?

[12AX7]

Ok, all this is over my head so i have questions. How do you measure voltage....from cathode to ground or heater to cathode? I get 160 to ground and 118 from heater to cathode. Looking at the electro harmonix sheet it says positive and negative voltages and shows "not more than" and "not less than" for the other. I have no idea what to do, what to shoot for as far as changing the elevated circuit or whether it will allow me to use a EH or any other tube based on the data sheet. And positive/negative? I get the same voltage whichever way i use the probes so i don't know if that means 2 different places to measure or what. I understand if you don't want to stoop to my level to explain it, but just saying i don't understand most of this.
http://drtube.com/datasheets/12ax7eh-ehx.pdf

{edit -- clarified URL -- PRR}

[Willabe]

It's not a big deal.

Like I said, I go with +70dcv to +80dcv for a DC standoff voltage. You just want to lessen the voltage stress between the tubes K and heater.

Most gain stages only have a volt or 2 on it's K so no problem. But with a CF you can have a 100+ dcv's on it's K as your measuring with your CF. Some new tubes don't seem to hold up well or at all with a high K dcv. NOS tubes seem to have been fine. 

Read this it will help you; 

http://www.valvewizard.co.uk/heater.html

[TitaniumValhalla]

The question i have is about several 12AX7;s that are not recommended for used in the CF position or they will fail. Had that happen in the past at least with the tung sols. ...

Got a link to the recommendation/warning?

Electro-Harmonix has the Reflektor factory make the "Tung Sol" tubes, as well as many of the other EH/Sovtek line of tubes.  EH's Tung Sol website includes data on the 12AX7 page that is really just a scan of the original Tung Sol 12AX7 data sheet.

And that original data sheet claims 180v as the max allowable heater-to-cathode voltage.

If you had failures with a new Tung Sol-ski at less than 180v heater-to-cathode, it would probably be an issue to take up with the retailer/distributor where you bought them.  The failure would likely have been due to tube defect or insufficient testing, and the same could happen for any brand of tube.

I was able to find this on New Sensor's website (who makes the Tung-Sols as well as EH and Sovteks). I had heard hearsay that all of New Sensor's spiral-filament tubes had this problem but no official confirmation from them. The Sovtek 12AX7WA/B/C are the only non-spiral-filament tubes they make as far as I know. This is probably the closest thing to an official confirmation from that most of their tubes do not meet those specs.

The Sovtek 12AX7WA/WB/WC series of tubes are general-purpose workhorse types that provide reliable performance in a variety of applications. These tubes are frequently found as OEM tubes in many different brands of amplifiers. The 12AX7WB has about 6% more gain than the 12AX7WA and a slightly darker sound signature, which is great for smoothing out harsh sounding amplifiers. The 12AX7WC has closely matched sections making it ideal for phase inverter circuits. The 12AX7WA/WB/WC types hold up better than other 12AX7 tubes in cathode follower circuits where the maximum cathode to heater rating of 100 volts is often exceeded.

from http://www.newsensor.com/faqs5.aspx

[Willabe]

How do you measure voltage....from cathode to ground or heater to cathode? I get 160 to ground and 118 from heater to cathode.

I'd just measure the K's dcv.

If your heater is sitting at 6.3acv and your K is sitting at 160dcv, then; 160 - 6.3 = 153.7.  So if the spec sheet says not more than 100v between the heater and K your 53.7v over. (Notice the spec sheet you posted a link for just says _ v _ no ac/dc, because the heater can be run with ac or dc.)

Yes tubes have been run over voltage for many decades and were just fine most of the time. But new tubes don't seem to take/handle over voltage as well as NOS tubes did/do.

[PRR]

> heater to cathode?

Cathode to heater center-tap.

> 160 - 6.3 = 153.7

Assuming *grounded* CT heater, there is 3.15V RMS each side. This is 4.4V peak (which is what matters). This can add or subtract. So the math is 160V + 4.4V, 165V.

And 160==165 for practical purpose.

If the heater CT is elevated to say +60V, with cathode at 160V, now we have 160V - 60V is 100V. Well, 96V-104V counting the small heater swing.

BUT that cathode also caries audio. Possibly 50V peak. So cathode swings say 110V to 210V. The worst-case is 210V-60V is 150V.

_I_ say the tubes should take 200V H-K. Macintosh worked a 12AX7 very hard. If it truly fails of H-K short, put in another. If that fails, get a different factory.

[Willabe]

Thanks PRR.   

[HotBluePlates]

Got a link to the recommendation/warning?

I was able to find this on New Sensor's website (who makes the Tung-Sols as well as EH and Sovteks). I had heard hearsay that all of New Sensor's spiral-filament tubes had this problem but no official confirmation from them. The Sovtek 12AX7WA/B/C are the only non-spiral-filament tubes they make as far as I know. ...
from http://www.newsensor.com/faqs5.aspx

Thank You for posting this.

I don't know what to make of it, as the Sovtek 12AX7WA/B/C variants are older than any of the other tubes they list.  Seems a mistake they would "go backward" with regard to specs, but who knows when you're taking Russian tube types & relabeling them with U.S. type numbers.

I agree with PRR.  I've got 2 McIntosh MC-30's which have a 12AX7 drive stage with -45v on the cathode and 445v on the plate (~490v plate-to-cathode on a tube rated for 300-330v max).

There's ratings, there's good engineering practice (keeping heater-to-cathode voltage small by elevating the heater reference), and there's practical reality.  I guess all 3 don't always show up to the party in audio...

[jjasilli]

Yes, thanks PRR for showing the math.  There's little to no hard info anywhere, including Radiotron 3 (haven't looked at 4 yet).  Meanwhile for decades Fender CF tonestack drivers have been putting 190V on the 12ax7 cathode with no heater elevation. 

[PRR]

> I don't know what to make of it

See the description of the KT150.

"glass envelope is a special egg shaped balloon... The fact that the glass envelope has no flat sides means the Tung-Sol KT150 has no problems with microphonics."

[PRR]

> showing the math.  There's little to no hard info

This is carpenter math. Joseph (Jesus' "dad") did it without a calculator. (Maybe without a zero.) He didn't have books and they probably didn't let him in the Nazareth scroll-library.

You want a myrrh-shelf 160 inches up. You go to buy shelf-braces but the first ones say "100 inch max". That won't work. Unless you start them from something 60 inches high. If that something also vibrates 4 inches, you may have to come up to 64 inch to ensure a 100 inch brace will reach. Or you can look for 200 inch braces.

[shooter]

they probably didn't let him in the Nazareth scroll-library.

Bureaucrats!

[Viktor86]

@ PRR

"with cathode at 160V"  "BUT that cathode also caries audio. Possibly 50V peak. So cathode swings say 110V to 210V."


If the Audio is 50V peak to peak, with cathode at 160VDC, wouldn't it swing from 135V to 185V?




Thank you!

[pdf64]

@ PRR

"with cathode at 160V"  "BUT that cathode also caries audio. Possibly 50V peak. So cathode swings say 110V to 210V."


If the Audio is 50V peak to peak, with cathode at 160VDC, wouldn't it swing from 135V to 185V?

Yes, that is explained in reply 18.

[tubeswell]

@Viktor86
This is a thread from 2017. You’re probably better off starting a new thread.

[Viktor86]

@ pdf64 and tubeswell


Thanks for replying


I have read reply 18 many times. I think PRR is a genius. I have found his posts very helpful over the years.


However.. It looks like he said, a 50v peak to peak signal at a 160vdc cathode, swings 110v to 210v.
Does 135v to 185v make more sense to you guys?


Thank you

[HotBluePlates]

It looks like he said, a 50v peak to peak signal at a 160vdc cathode, swings 110v to 210v.
Does 135v to 185v make more sense to you guys?

Re-read the post:  "Possibly 50V peak."

That's 100v peak-to-peak.  But it's simpler; 50v in each direction, so 110v and 210v.

[Viktor86]

Thanks HotBluePlates


That makes sense. I misunderstood PRR.
I’m finding out about peak voltage (vs peak to peak voltage)
So 50V peak = 100V peak to peak


100v peak to peak may be the worst case scenario. It may be what some high gain amps with a high B+ can swing when pushed all the way.

Luckily, mine swings 55v max.
With a cathode at 130vdc, and heaters dc referenced at over 60v, I can use any 12ax7 rated at 100v for Vhk.