Yet another 12B4 line stage, or is the 12B4 better than the Grounded Grid..... - Page 12 - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 21st December 2005, 06:00 AM   #111
diyAudio Member
 
sunrise's Avatar
 
Join Date: May 2005
Location: delta quadrant
Send a message via ICQ to sunrise
Dear Mark,
I was running my 12b4 in grounded cathode connection on arround 10mA. It was o.k. and was sounding nice. Than I decided to try LM317 but because of the higher current and my power supply configurationmy resistors in my power supply lower my high voltage and to try this circuit this was the easiest way.... my tube is running on 26mA now but i will try few things - first to lower the current and second to bypass the capacitor in the cathode.....
Also i will try to manage to have my high voltage higher to try it this way....

Pedroskova - yes, that is what i was asking here... if this little bug that is down there causing the problems that i have... but i am not sure if i use good quality resistor would it be much better than a normal no-name res. i simply have to try it..... and here i still have the problem of the lower high voltage if i would like to drive the tube arround 30mA.....

well according to datasheets the tube (as in sch i attached few mails ago) is running in a very linear mode.....but it can't swing the best she could.....
__________________
when you burn it - burn it good
  Reply With Quote
Old 22nd December 2005, 12:02 AM   #112
diyAudio Member
 
Brian Beck's Avatar
 
Join Date: Dec 2004
Location: South Florida, USA
Quote:
Personally, I'd lose the silicon hanging onto your cathode and just go with a traditional grounded cathode stage. It's could be causing your treble problems.
(It's hard to keep up with the pace of some of these threads...) Anyway:

I don't think so. I'm as much against silicon ruining tube circuits as anyone. The reason I originally showed this CCS-in-the-cathode circuit was just to maintain the 12B4 at a constant 30mA, completely independent of plate voltage drifts, B+ drifts, cathode (Vgk) drifts and tube aging. As a side benefit, the RC time constant determining the bass roll-off at the cathode is larger (longer) using the CCS, making for a lower -3dB roll-off frequency for the same size cathode bypass cap(s). All for 50 cents - maybe cheaper than a fancy cathode resistor. The LM317 must be bypassed by large paralleled caps. You can choose whatever caps your heart desires here, as fancy or using as many in staggered sizes as you might wish. At any audio frequency, the caps will present a much, much smaller impedance (thousands or millions of times lower) to ground than the LM317, effectively shorting it out. Even if it is not so good a CCS above a few hundred hertz (and it's really not so good up there), the caps will dominate the sonic picture completely. This is a TOTALLY different situation than when using CCSs as plate loads. I proceed with greater caution there. There are no caps to ďhide underĒ for that CCS position. I generally donít like silicon on the sensitive plate at all. Gary Pimmís CCS designs may be the exception.

The design posted by Sunrise really needs a lot more plate voltage, IMO. Also, the supply must be quite clean.
__________________
Brian
  Reply With Quote
Old 22nd December 2005, 01:02 AM   #113
diyAudio Member
 
Join Date: Sep 2001
Location: Salt Lake City
Brian,

That probably explains the definate for the better change in the low end that I experienced after changing to the LM317 CSS. The multiple caps are a definate must and alot more experimentation will be done by me in this regards as to the audible differences in paralleling different caps.

As for the supply end I feel that a good Shunt type SS supply like Allen Wright's followed by a pair of OD3's similar to but not exactly the same as pedroskova's would be a good comprimise place to start on a high quality(and still size managable) SS supply that would sound dynamite. The SS regulator would make the supply ultra stable, clean and very would have a low impedance over the audio range we deal with....the gas regulators acting as a second shunt regulator would serve to improve on the SS's delivery system.

Mark
  Reply With Quote
Old 22nd December 2005, 05:23 AM   #114
diyAudio Member
 
Join Date: Oct 2005
I'm collecting parts to build one of these too. I was planning to run the tube at 30 mA with 160 V on the plate, and a CCS is a definite. (Any comments on these choices?)

Unfortunately, I don't know if I should tie the CCS to the cathode or the plate. From Brian's post I understand that the cathode is the more forgiving of the two choices. But, what are the advantages of placing it above the plate? Does the plate option reduce distortion?

I'm trying to understand the realtive merits of these two choices. Any comments would be greatly appreciated!

Thanks to all.
  Reply With Quote
Old 22nd December 2005, 05:47 AM   #115
diyAudio Member
 
sparkle's Avatar
 
Join Date: Oct 2003
Location: middle of nowhere...
o.k. in another words i have two choices.... one - totally reconfigure the power supply and buy another transformer with higher voltage on secondary or, two, lower the plate current and try the circuit when running arround 15mA.
Anyway i will have to bypass the cathode cap. good with a smaller one capacitor Mcap or something.....
i have 47kohm in my power amplifier (solid state transistor) - not so hard to drive but needs a quality preamplifier to drive it that has a lot of dinamics and everything
regards
sunny
  Reply With Quote
Old 22nd December 2005, 12:21 PM   #116
diyAudio Member
 
Brian Beck's Avatar
 
Join Date: Dec 2004
Location: South Florida, USA
Quote:
Unfortunately, I don't know if I should tie the CCS to the cathode or the plate. From Brian's post I understand that the cathode is the more forgiving of the two choices. But, what are the advantages of placing it above the plate? Does the plate option reduce distortion?
The two options are not an ďapples-to-applesĒ comparison. A CCS in the cathode serves a different, more modest, and more easily implemented function than one in the plate.

A CCS in the plate circuit *CAN* improve linearity, although it's not easily done, IMO. The problem is that CCSs in the plate circuit can be heard. Any non-linearity in the CCS's effective impedance with voltage swing, and especially any changes in the CCS's capacitance with voltage swing, will impart distortion and ruin the nice inherent qualities of a tube like the 12B4. To do a plate CCS right requires many components, not just a modest LM317 (which canít handle the voltages anyway). I suggest searching for Gary Pimm's fine plate CCS designs. While I havenít built them according to his exact designs, others whom I respect have listened to his CCSs critically and have found them to be excellent. Gary went to extraordinary lengths to reduce capacitance to fractional-picofarad level, and to increase effective resistance to many giga-ohms. But the result is high complexity.

For the original 12B4 circuit I posted, I was not even considering a plate CCS: I used a wire-wound resistor as a plate load. I happened to use 6.2Kohm with a 320V B+ supply that resulted in about 140 volts on the 12B4 plate (to ground, not to cathode). The effective plate resistance of a 12B4 is about 1000 ohms when 30mA is passed through it. So I chose a plate resistor about 6 times higher than rp. This is a simple design and with a good wire-wound can sound excellent without the worries and complexity of a plate CCS. I used a rectangular wire-wound of no particular distinction. A Mills 12Watt wire-wound would probably be best. A fanatic could always increase the B+ to higher voltages, and therefore be able to increase the plate load resistor to larger values, although with diminishing returns. For example if you went to 450 volts on the B+, you could increase the plate wire-wound to 10Kohm, but it would have to be a 20 watt item or so (and now you have higher power supply cap ratings to worry about). Maybe itís worth it, but probably not. And again, the LM317 CCS in the cathode circuit is hiding beneath a parallel set of caps, so all it does is to ďregulateĒ the 12B4ís cathode/plate *DC* current to 30mA against drift (and with a slight bass extension I mentioned earlier). If it isnít the greatest CCS in the audio band, it doesnít matter because the bypass caps will determine the resultant sound. No more than that. And with just ~50 cents of parts.

If you decide to do a plate CCS, donít half-bake it; do it right. And again see Gary Pimmís site. Another plate option is always a plate choke, although at 30mA of plate current it will not be easy to get enough inductance for low bass response.
__________________
Brian
  Reply With Quote
Old 22nd December 2005, 05:12 PM   #117
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
In the ideal (textbook) world, a perfect triode would generate the lowest distortion and highest gain when its plate load is an ideal current source. Unfortunately, our parts are made in the real world. I am a big fan of using CCS plate loads for triodes, in order to get as close to ideal results as possible. The CCS plate load will allow the triode to operate more linearly than a resistor if there is no AC load on the tube. Unfortunately you can't get any signal out of the tube without puting a load on it. There are many ways to buffer this load to minimize it. The Mu follower, SRPP, and Gary Pimms active load CCS's are a few. I use a simple mosfet follower in my designs.

With the theory out of the way, it has been found that the "best sound" does not always correlate with the lowest distortion. And some tubes respond better to CCS plate loads than others. I have not experimented with the 12B4 (and its relatives) yet, but this thread has sparked my interest, and I need a line stage with moderate gain, so I am going to build one of these. It is just my hunch, but I think that this design would be better off with the LM317 circuit in the cathode and a resistor load.

I know from experience that you can't do both. In a blonde moment I tried putting a CCS chip (IXYS 10M45) in the cathode and plate circuit of a 45. After I scratched my head wondering how I built a single ended flip flop, I remembered that basic electronics class from 40 years ago and a guy named Kirchoff.
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 22nd December 2005, 05:34 PM   #118
VT67 is offline VT67  Belgium
diyAudio Member
 
VT67's Avatar
 
Join Date: Jun 2004
Location: Belgium
Tubelab

Sorry for going off-topic for a moment. Two weeks ago I sent you an email via the email address of your website, did that came through?

Regards
  Reply With Quote
Old 22nd December 2005, 06:19 PM   #119
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
I received several e-mails in the past two weeks. Computer at home is dead, I have got a new one, but everything is not completely restored yet.

I have been participating in the forums from my computer at work, or in a hotel room on a recent trip. They forbid any "activities related to secondary employment" at work, so I can't respond to the e-mail (computer use is monitored). I will be off of work for the rest of the year as of tomorrow, so I won't be on the forums until my computer is restored at home.

There will be a major update to my web site in January. The SE amp PC boards will be available then. The entire comstruction manual will be posted also. I don't plan to offer parts kits at this time (maybe later) due to the fact that everyone wants something different. It would be impossible to keep parts in inventory and sell them in small volume without charging more than they are worth.

I designed the amp using parts that were available from Digi-Key, because they ship world wide. I am in the US so I can't speak for their service outside the USA. If this is a big issue, let me know.

I will answer all e-mail to the Tubelab site as soon as my computer is restored. I will be out of town until Tuesday, so it will be next week.
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 22nd December 2005, 06:23 PM   #120
VT67 is offline VT67  Belgium
diyAudio Member
 
VT67's Avatar
 
Join Date: Jun 2004
Location: Belgium
Thanks for the info
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 12:37 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2