Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum

Message Sort: Post Order or Asylum Reverse Threaded

I don't get this...

62.163.171.12
Posted on October 16, 2014 at 03:18:42
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 HI,

I'm trying to establish how much current is flowing trough a WE437A tube I use in a phono stage.

My B+ is 184V. On top of the tube ithe primary of an 1:1 interstage with a DCR of 1385 ohm.

I measure 167V on the plat of the tube. This means a voltage drop of 17V over 1385 ohms, so about 12mA,

The tube is biased via filament bias. I measure 2.1V bias on the cathode/positive leg of the heater. So the grid is 2.1V negative to the cathode, so -2.1V bias.

Eying the WE datasheet, 167V on the plate and -2.1V bias should give me 50mA of current (a bit too much I know).

I tried this with two pairs of tubes, with roughly the same results.

The tube swings *very* little voltage at this point btiw.. its sounds great...

Anybody?

 

Hide full thread outline!
    ...
RE: I don't get this..., posted on October 16, 2014 at 03:54:18
Al Noakes
 High gm valves test all over the place and rarely meet the published specs from my own experience

'The tube is biased via filament bias'

Why not use stacked diodes in the cathode circuit ? I think you will be making the stage a lot more noisy than it should be using the heater supply to provide the bias .

Al

 

Al is on the money ..., posted on October 16, 2014 at 04:15:22
Naz
Audiophile

Posts: 2184
Location: Sydney
Joined: September 2, 2005
 on all counts.

A pair of stacked SiC Schottkies in the cathode will give you around 1.7V at that current and around 15mA. Moreover they'll sound better and quieter IMO.

Naz

 

Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 05:00:58
Chip647
Audiophile

Posts: 2652
Location: The South
Joined: December 24, 2012
 The heater and the cathode are separate in the WE437A.

 

RE: I don't get this..., posted on October 16, 2014 at 06:12:33
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 Hi,

I generally do not like any sand in my amplifier, especially not in the current loops.

I'm not using just 'any' resistor as bias resistor...(see pic). These are silver wire wounds made by Audio Consulting (Switzerland).

Yes, it generates still some noise, but my RIAA it's more quiet than the $250K Kondo M10 phono I once had ;). Noise of my record are generally higher ;)

NC

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 06:16:15
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 Hi,

Actually they do have a filament, but most of the time its called a 'heater' nowadays on indirectly heated tubes. On the old datasheets you still see Ef mentioned (from Filament).

So you just tie top of the filament resistor to both the cathode and one end of the heater/filament. then the cathode will become positive wrt the grid.

NC

 

RE: Al is on the money ..., posted on October 16, 2014 at 06:23:11
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005



Hi,

I generally do not like sand in my amps, especially not in the current loops.

I'm not using just 'any' filament bias resistors (see pic). These are silver wire wounds from AudioConsulting (Switzerland) I use for the first stage. I use a Kiwame resistor in the second stage.

Yes, its still a bit noisy, but not more so then the $250K Kondo M10 I once had. My records are generally noisier ;)

Quick question. When using diodes in the cathode... Am I correct that this will not influence the Zout of the stage? That would be a great benefit compared with cathode/auto bias.

Thanks,
NC

 

RE: I don't get this..., posted on October 16, 2014 at 06:38:51
Al Noakes
 That is going to be a huge waste of money just like anything from Audio Consulting . Don't they sell knobs that cost $250 each ? Are you a sandophobe as you used sand and not liked it or just averse to using sand devices full-stop ? I'm no purist and tend to use whatever sounds best :) I'd suggest trying some schottkys in the cathode , it will be cheap and easy to find out if these devices as cathode bias are to your liking :)

Al

 

The money is already spend :), posted on October 16, 2014 at 06:52:24
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
Its soo lovely to see that everyone has an opinion about everything in this forum :)

Did you try the knobs? ;) (me neither).

But I'm very happy with using filaments bias. No sand, keeps Zout low, no need for a cathode bypass cap etc. Only adds a small high quality resistor (in my case) in the signal path.

Enjoying the music...

NC

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 06:53:38
Chip647
Audiophile

Posts: 2652
Location: The South
Joined: December 24, 2012
 The WE datasheet I found said that the tube was indirectly heated. This means that the heater and cathode are electrically isolated. The tube has 2 heater pins and 2 cathode pins. The max heater to cathode voltage differential is 50 volts. Are you hooking the cathode to the heater in series or something?

I have implemented filament bias and I like it, but only with filamentary cathode tubes.

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 06:59:29
jazbo8
Audiophile

Posts: 70
Location: Europe
Joined: January 16, 2012
 Agree with Chip, isn't this an in-diectly heated tube?

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 07:02:39
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 Hi,

I do it just like shown in the link. I connect the top of the resistor to one side of the heater AND I conned this side to the cathode.

How can you say you like filament bias only on DHT's while you obviously never tried it on indirectly heated tubes?

NC

 

Yes, and I never said it was not..., posted on October 16, 2014 at 07:04:04
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 I said it has a heater...

 

RE: The money is already spend :), posted on October 16, 2014 at 07:10:11
Lew
Audiophile

Posts: 10912
Location: Bethesda, Maryland
Joined: December 11, 2000
 It's not the boutique resistor I would worry about so much as any noise that is inherent in your filament supply. If you have time on your hands one day, just try using a Schottky to bias the cathode. I got the idea from Naz and others like Al, and it does work a treat; there will be LESS noise, not more noise, sand-o-phobia notwithstanding. (By the way, a Schottky is not made of sand, anyway.)

 

RE: The money is already spend :), posted on October 16, 2014 at 07:52:30
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005

Well, its quiet enough for me, but I will try it one day to see if it will be better sounding.

Still not sure how the diode will affect the Zout though...

Thanks,
NC

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 08:02:13
Chip647
Audiophile

Posts: 2652
Location: The South
Joined: December 24, 2012
 Confused initially because indirectly heated tubes do not have a filamentary cathode. You are right, I have never used the heater as a bias resistor to raise the cathode potential.

 

RE: I don't get this..., posted on October 16, 2014 at 08:23:11
Tre'
Industry Professional

Posts: 17305
Location: So. Cal.
Joined: February 9, 2002
 "I generally do not like any sand in my amplifier, especially not in the current loops."

But isn't there "sand" in your heater/filament supply?

I guess it's "shunted" by a 3 ohm resistor so maybe my question is moot.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

Forget about the SAND guys... I was speaking metaphorically!! NT, posted on October 16, 2014 at 08:33:05
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005

 

Wrong reply from me Naz..., posted on October 16, 2014 at 10:06:55
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 Got confused with browser tabs...

What i wanted to ask you, can you please explain how this can be "on the money".

My datasheets tells me I should expect over 45 mA when using 165V on the plate with -2.1 bias.

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 10:39:45
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
 I think you need to understand why people use filament bias before you look at an indirectly heated tube and pronounce that it's a good idea.

In your 437A phono preamp, you have the advantage that noise in the heater supply has some isolation from the cathode. If you try to integrate the heater supply into the bias supply, you will introduce a lot of noise.

You cannot filament bias an indirectly heated tube because it has no filament, period.

The arrangement you are describing is more like tapping the heater supply for a fixed bias voltage.

 

You don't use the heater as bias resistor.., posted on October 16, 2014 at 10:41:53
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
The current through the heater in combination with a separate bias resistor from heater to ground creates a node with a certain positive voltage. By connecting the cathode to this point, the cathode is raised to the same voltage.

NC

 

RE: Wrong reply from me Naz..., posted on October 16, 2014 at 10:46:38
Tre'
Industry Professional

Posts: 17305
Location: So. Cal.
Joined: February 9, 2002
 "My datasheets tells me I should expect over 45 mA when using 165V on the plate with -2.1 bias."

Looking at the plate curves I get about 160ma. Is that right?

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 10:50:52
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 I dont' agree...

See my response below.

I filament bias because

1) I don't want to use cathode bypass caps and
2) I don't want to raise the Zout of the stage. A cathode resistor at this place would be much larger, adding to more ohms to the output Z. I cannot have that because its driving a 10K LCR directly, so I want Zout under 1K. The current through the heater is much larger then through the tube, so the resistor value can be lower, and so Zout will be lower.

As said again and again, the noise is not a problem... I can hear it only when he volume pot is over 80% and its not related to the signal.

The Schottky diode option is new for me. I might try this once as said below. I'm sure it will work of course, but I doubt it will sound better. But again, I will try.

Its quitter then several commercial phone stages I've heard.

NC

 

RE: You don't use the heater as bias resistor.., posted on October 16, 2014 at 10:52:07
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
 That's simply a fixed bias supply.

You could accomplish the same task with a resistor and a zener doide of your desired bias voltage, without stacking the heater on top of a resistor.

If you are transitioning from the "normal" resistor/capacitor combination to something with vastly lower impedance and better noise performance, diodes are the answer.

Otherwise, you will have a hell of a time getting your 8.2V supply quiet enough, especially in a phono preamp.

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 10:54:46
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
 It simply isn't filament bias, but rather a fixed bias supply.

If you have capacitors in the DC supply feeding the tube/resistor, then you haven't exactly accomplished your goal of not having a cathode bypass cap.

This is sort of like saying that you have a regulated screen grid on a triode. A triode doesn't have a screen grid, just as an indirectly heated tube has no filament.

(Your plate voltage shouldn't be that far off IMO by-the-way)

 

RE: Wrong reply from me Naz..., posted on October 16, 2014 at 10:55:01
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 When I look at the I/V curves, I see about 50mA

Plate Voltage = 167
Ec = -2.1V



 

RE: Al is on the money ..., posted on October 16, 2014 at 10:56:37
jazbo8
Audiophile

Posts: 70
Location: Europe
Joined: January 16, 2012
 "I'm not using just 'any' filament bias resistors (see pic). These are silver wire wounds from AudioConsulting (Switzerland) I use for the first stage."

So the filament bias resistor is 3R only? Or are there other resistors in series with it?

 

I read it wrong, I'm sorry. You're right. , posted on October 16, 2014 at 11:00:13
Tre'
Industry Professional

Posts: 17305
Location: So. Cal.
Joined: February 9, 2002
 The decimal point is only showing up, for me, on the -3.0 and -5.0

That was silly of me.

tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Al is on the money ..., posted on October 16, 2014 at 11:05:23
Tre'
Industry Professional

Posts: 17305
Location: So. Cal.
Joined: February 9, 2002
 The whole point of filament bias is that the plate current AND the filament (heater in this case) current is flowing through the resistor so the resistor does not need to be very large so very little local current feedback will exist and there will not be a need to bypass it.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Quick question: How do you filament bias a tube that does not have a filament?, posted on October 16, 2014 at 11:06:00
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005

Its not a fixed bias supply.

My method is the same as the method called filament bias, I only use the heater instead of the filament (which I'm sure is also called a filament in some books). The 'old' reference is Ef (also for indirectly heated tubes).

However, you can call it anything you want, I'm fine with that.

The heater voltage is supplied by a VCCS.

NC



 

NP, but are high mu tubes really that out of spec??, posted on October 16, 2014 at 11:10:19
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 I mean I tested with 4 tubes. The tubes all give the same values more or less (12 - 15mA).

But WAY out of spec... like completely.

12mA @ 167V seems not a very linear region to me. I aimed for 25mA, but got a bit less bias than expected. Apparently the heater draws a bit less current then the data sheet says.

NC

 

RE: The money is already spend :), posted on October 16, 2014 at 11:18:14
Al Noakes
 Yes I have an opinion :) The heater supply in filament bias places any ripple or noise onto the grid as it is referenced to the cathode , I also believe the rejection to common mode noise is also reduced . There is sand in that supply as it must be DC . Seriously , do yourself a favour and rip that bias scheme out . By all means use DC on the heater but get that heater supply out of the signal path . It should make your amp quieter and probably also sound better .

Filament bias is best suited to small , low-current DHT fils , but for me , has very limited appeal and I don't use it anymore . I went through a phase of using filament bias on everything a few years ago including bastard-filament valves such as 3C24 and 826 . It's bulky , can be noisy and adds a sound of it's own . Schottky diodes or LEDs provide a workable , cheaper alternative , to be honest I also believe film caps/resistors sound better

Al

 

RE: Al is on the money ..., posted on October 16, 2014 at 11:25:48
jazbo8
Audiophile

Posts: 70
Location: Europe
Joined: January 16, 2012
 Just checking since the numbers provided do not make sense.

 

RE: The money is already spend :), posted on October 16, 2014 at 11:27:10
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 Hi Al,

Thanks for your opinion Al :)

So, to be sure... diodes will not raise the Zout of the stage like an Rk does?

I use a VCCS... Lotsa sand, but no idea about what is and what's not in the input current loop.

Another problem I see with diodes.... It looks difficult to get the exact amount of bias you need. Especially with high mu tubes.

NC

 

It IS quiet enough. I build it already and listening to it! (nt)., posted on October 16, 2014 at 11:36:51
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 .

 

I think they can be. (nt), posted on October 16, 2014 at 11:54:45
Tre'
Industry Professional

Posts: 17305
Location: So. Cal.
Joined: February 9, 2002
 .
Have Fun and Enjoy the Music
"Still Working the Problem"

 

I think I solved the mystery, now verifying... (nt), posted on October 16, 2014 at 11:56:41
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 .

 

RE: Al is on the money ..., posted on October 16, 2014 at 12:00:05
twystd
Audiophile

Posts: 2723
Location: Austin,Texas
Joined: December 9, 2000
 I have tried diode bias on a CCS parafeed line stage, and am very pleased with the results. Not sure what effect that diode bias has on output impedance with a conventional gain stage. With parafeed, you can take advantage of the Western Electric output topology (parafeed cap returning to the top of the cathode), which I believe, would not increase the output impedance of the stage, since the AC signal path is restricted to just the tube, parafeed cap, and the OPT.

The benefit is that you can eliminate the cathode bypass cap. I know, we've just substituted a parafeed cap for a cathode bypass cap, however the parafeed cap is much smaller, and can therefore be of higher quality. I preferred this over battery bias as well.

twystd

 

RE: The money is already spend :), posted on October 16, 2014 at 12:05:49
Al Noakes
 You can pad down the diodes with small value resistors to get your required bias point .

Why worry about Rk ? Are you driving something that needs a low impedence such as an LCR filter ?

Al

 

Yes, a 10K LCR... , posted on October 16, 2014 at 12:08:58
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 But thats only in the first stage. The second stage drives an IT, which drives an TVC.

NC

 

RE: Yes, a 10K LCR... , posted on October 16, 2014 at 12:26:06
Al Noakes
 I'd say : try it ! Should be easy to revert back if you don't like what it does . I'd suggest on the input stage first

Al

 

MYSTERY SOLVED! thanks all..., posted on October 16, 2014 at 13:28:55
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 HI,

The mystery is solved.

When adjusting the heater voltage (to be 6.3V), I wrongly measured between + and ground (since this was easier to do).

Of course I need to measure between + and the top of the filament bias resistor.

Now the filaments where at 4.2 volts (6.3 - 2.1 bias) .

So less electrons emitted as designed, so less current measured.

I needed to raise the voltage to 8.8 volts to get 6.3 volts between both legs of the heater (note, the WE437A stage has a 5.6R filament resistor, not the 3R resistor shown).

Now it makes more sense.

VCCS out = 8.8V
Bias = -2.5V
Heater voltage is 6.3V

B+ = 185V
Va = 160V

So, 25V drop over 1385 ohms is 17mA

Eyeballing the datasheet I should get over 20mA, but its close enough.

NC

 

Not so easy to try ;), posted on October 16, 2014 at 13:39:50
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 Did you see my RIAA stage?

Its quite a bit of work, but I will try it once... Thanks for the tip!

(still curious about the theory of the diodes influencing Zout though)

NC








 

Btw... it *sounded* great with filaments at such a low voltage..., posted on October 16, 2014 at 13:44:22
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
But really....

I cannot listen now with the new voltages since I have sort of an agreement with my neighbours about listening hours...

But jeez, who needs 6.3V if 4.2V sounds so great...

 

Sorry *heaters*... (nt), posted on October 16, 2014 at 14:11:42
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 .

 

I generally don't like sand either but ..., posted on October 16, 2014 at 14:47:32
Naz
Audiophile

Posts: 2184
Location: Sydney
Joined: September 2, 2005
 for biasing gain stages such as this I've found no better way than Schottky diodes, SiC providing greater voltage drop. They sound better than anything I've tried, particularly for phono applications and I've spent serious time evaluating the options. Although I've not tried every other diode possible Schottky diodes generally sound way better than silicon diodes to my ears.

Since they have very little dynamic resistance local nfb is minimal so they don't have to be bypassed. Therefore OP impedance is almost as good as running a piece of wire in the cathode circuit.

Serious resistors you are using btw, very impressive looking!

Cheers,
Naz

 

RE: You don't use the heater as bias resistor.., posted on October 16, 2014 at 18:44:41
Chip647
Audiophile

Posts: 2652
Location: The South
Joined: December 24, 2012
 I can see the benefits for a filament bias more clearly than heater current through a resistor bias as there are other, better options for biasing an indirectly heated tube, IMHO. I am glad that you enjoy it, however. Measure the current using the plate resistor.

 

OP Impedance ..., posted on October 17, 2014 at 04:34:52
Naz
Audiophile

Posts: 2184
Location: Sydney
Joined: September 2, 2005
 It takes an appreciable amount of resistance in the cathode to affect OP impedance significantly.

The equiv circuit of diode is a voltage source (by definition zero impedance) with a series resistor, which is typically only a few ohms at the milliamps of current typical of a gain stage.

Seriously, try Schottky, they sound great and are dead quiet. SiC will give you around 0.8V and standard Schottky runs about 0.2V in these applications so a combination will give you flexibility up to a few volts.

Naz

 

Zout will not be negatively affected using Schottky bias (nt), posted on October 17, 2014 at 04:39:07
Naz
Audiophile

Posts: 2184
Location: Sydney
Joined: September 2, 2005
 Naz

 

Ah, thanks... Any clue about how/how much/the theory (always curious).. (nt), posted on October 17, 2014 at 04:41:14
Nickel Core
Audiophile

Posts: 870
Joined: February 11, 2005
 .

 

that sure has been my experience with octals... perhaps worse with 9 pin because of closer quarters?~nT, posted on October 17, 2014 at 06:05:04
Cleantimestream
Audiophile

Posts: 7551
Location: Kentucky
Joined: June 30, 2005
 ~!
The Mind has No Firewall~ U.S. Army War College.

 

Page processed in 0.029 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]