12BZ7

[Mosquito]

Hi everyone!
I have a bunch of 12BZ7 tubes, and want to to make good use of them. The general opinion I heard was "every section is like two // 12AX7 triodes"
I wonder if someone actually used it, and found some sweet spot, I mean like the "200V on the plate at 3 or 4 mA" preferred by 12AT7s.. Of course I can experiment and found by myself, and I' ll do, but I found interesting to hear about some others experiences
Cheers
J

 

[rongon]

I know that some of the more experienced builders here have stated that 12BZ7 is not very linear, so don't use it. It's probably not horrendous, so might be useful. But if it was good for use in a sensitive spot, like the input stage of a phono preamp, you'd think somebody would have used it by now.

Come to think of it, higher distortion in the input stage of a phono preamp wouldn't be a deal killer there, as signal levels are so low. I wonder if the problem is crippingly high input (Miller) capacitance?
--

100k loadline with 450V B+ doesn't look inspiring...
--

 

[SY]

It's touted to be like two 12AX7s, but the linearity is decidedly poorer, while still being at least as difficult to use properly.

 

[jazbo8]

Great for guitar amp though...

 

[20to20]

I know that some of the more experienced builders here have stated that 12BZ7 is not very linear, so don't use it. It's probably not horrendous, so might be useful. But if it was good for use in a sensitive spot, like the input stage of a phono preamp, you'd think somebody would have used it by now.

Come to think of it, higher distortion in the input stage of a phono preamp wouldn't be a deal killer there, as signal levels are so low. I wonder if the problem is crippingly high input (Miller) capacitance?
--

100k loadline with 450V B+ doesn't look inspiring...
--

It looks fantastic for a 1v. input. This tube is meant to drive something like a 6K6, from the same source that would normally use an AX7. The graph of paralleled AX7's would probably tighten up, too. So you get 1/2 the Ra and 2x the Gm. in one tube.... You have to use it for what it was designed for.

 

[GoatGuy]

Its good for guitar amplifiers, for sure. In fact, one can just "drop it in" an AX / AT front end, and immediately hear a marked difference. Not I think its fatter, but dâmn!

Anyway, it also works pretty well in audiophile amps so long as it is not capacitor-cathode-bypassed for nominal biasing. Cathode local feedback works well to tame its modest non-linearities. Very well, actually. Yes, some gain is given up.

GoatGuy

 

[Mosquito]

Thanks my friends, very useful info, and advice from all of You.

This tube is meant to drive something like a 6K6, ...You have to use it for what it was designed for.

I think I will use them the to make the first stages of a PP, not hi fi maybe, but a thing with the D. Self "niceness" button always on

Cathode local feedback works well to tame its modest non-linearities. Very well, actually. Yes, some gain is given up.
GoatGuy

You mean local feedback cause the unbypassed cathode resistor, or some other circuit arrangement?

I do not want to mess with the Akira *ax7 model mainly because I do not even know where to start, so do You think that (for simulation purposes) 2 // sections of 12AXZ7 are a good enough aproximation?

 

[wa2ise]

2 // sections of 12AX7 are a good enough aproximation?

That should do it as well as an actual model of the 12BZ7.

 

[DF96]

The graph of paralleled AX7's would probably tighten up, too.

Putting two identical valves in parallel does not change the shape of the graph. To get that you would need to parallel two valves with significantly different mu.

do You think that (for simulation purposes) 2 // sections of 12AXZ7 are a good enough aproximation?

Yes, probably, as long as you are not interested in distortion. To get reasonable distortion figures from a simulation you need good valve models. Modelling a non-linear valve using a model intended for a much more linear valve guarantees that distortion figures must be ignored.

 

[Mosquito]

Yes, probably, as long as you are not interested in distortion.

thanks, in fact this will be the case here, an amusing non hi fi "niceness" generator
something of the like

 

[Sheldon]

It's touted to be like two 12AX7s, but the linearity is decidedly poorer, while still being at least as difficult to use properly.

I had an amp that I tried them with and compared with 12AX7. I didn't change resistor values, as they worked out reasonably for both tubes. Not really apples to apples. But I could get some of the BZ'z to give the same measured output distortion pattern as the AX's. I did find they they varied quite a bit more from tube to tube.

Sheldon

 

[Mosquito]

Well, this is what I come out with, just in case someone wants to check the sym..

 

[cogsncogs]

Well, this is what I come out with, just in case someone wants to check the sym..

That 12BZ7 model is way off. Try this one... Not perfect, but much better.
Grid current not modeled. Diode is added only to help with convergance, such as a cathode follower that's direct coupled to previous stage.

* 12BZ7_Syl LTSpice model
* Koren model (5 parameters): mean fit error 0.128482mA
* traced by Wayne Clay using Curve Captor v0.91 and
* Engauge Digitizer on 07/12/2016 from Sylvania datasheet
.subckt 12BZ7_Syl  P G K
    Bp  P K  I=(0.1187102759m)*uramp(V(P,K)*ln(1.0+exp((1.68598287)+(1.68598287)*(216.5801898)*V(G,K)/sqrt((5.044806155k)+(V(P,K))**2)))/(1.68598287))**(0.8527480618)
  Cgp G P 3.2p ; 0.7p added (2.5p)
  Cgk G K 7.2p ; 0.7p added (6.5p)
  Cpk P K 0.9p ; 0.2p added (0.7p)
  Rpk P K 1.0G ; to avoid floating nodes in mu-follower
  d3  G K dx1
.model dx1 d(is=1n rs=2k cjo=1pf N=1.5 tt=1n)
.ends 12BZ7_Syl

 

[Mosquito]

cogsncogs, thanks a lot. I ll try it right now.
Cheers
J.
http://www.diyaudio.com/forums/members/cogsncogs.html

The new spice model shared by cogsncogs yield to less power and more distortion in this circuit, all other things being the same. This was to be expected due the too optimistic results using the old model

 

[rongon]

I wonder if one took Wayne's 12AX7 model and paralleled it with Ayumi's 12AX7 model--in other words, create a model of a mismatched pair of 12AX7's in parallel--would you get something similar to a 12BZ7?
--

 

[cogsncogs]

I wonder if one took Wayne's 12AX7 model and paralleled it with Ayumi's 12AX7 model--in other words, create a model of a mismatched pair of 12AX7's in parallel--would you get something similar to a 12BZ7?
--

I ran your idea through Micro-cap and the result is below in the screenshot. Nowhere near the published curves, which look a little 'flaky' to me.

 

[rongon]

That suggests that the 12BZ7 does not equal two 12AX7 sections in parallel. It's a different tube with different characteristics.

I've tried a couple of 12BZ7 tubes, once in a Fender guitar amp and once as the LTP phase splitter/driver for a little PP 6CK4 amp (old JC Morrison design). In neither case did I think 12BZ7 was anything at all special. I think it might be useful in the LTP phase splitter stage of a Fender guitar amp, in place of a 12AT7. You'd get more gain that way, and it would stay at a better operating point than would a 12AX7 in that place.

The published 12BZ7 curves look really awful at high plate voltages with low-ish plate current. Running it with low plate voltage and high current would force you to use low grid-cathode voltages too, which risks inducing grid current. The shape of the curve for Ec = 0V makes me think grid current would be a problem with grid voltages close to 0V.

Ep 155V, Ec -1.2V, Ip 2.4mA looks like a good spot, but only if there's not a lot of voltage swing needed.
--

 

[cogsncogs]

Although it's not mentioned in the datasheet, but by just looking at the curves it looks like it may have a somewhat semi-remote cutoff characteristic. After all it was designed for sync separator/amplifier-clipper in television receivers. It mentions it's use as a general purpose AF amplifier, but that may be because of the high-mu.

 

[Mosquito]

Thanks Cogsncogs and rongon, clearly it is not a tube you want to get out and buy, is just the case that I already have a discrete quantity of them, because the mentioned Morrison amp, and because they were cheap..

 

[rifraf50]

12bz7 is ok

The 12bz7 is a worthwhile tube to use in head units for guitar work …..I have a number of RCA tubes and find they are more focused in the midrange area than standard 12ax7, perhaps warmer ?....they work well for jazz styles as well as not too aggressive rock styles ….I wouldn't use them in a combo because vibration may be an issue down the track

they draw more heater current than a 12ax7....600ma @ 6v ???

you would need to get a taller shield / retainer for it but as far as micrphonics / hum go , I've had no issues with any of them yet.

if you are going to use one in a device that is powered by a wall wart I would urge caution . the extra current needed to supply the tube may starve your device , burn your tranny , etc etc