Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

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

SEP using 6P30B and 5744WB/6C7B
SEP using 6P30B and 5744WB/6C7B
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 12th February 2019, 06:19 PM   #11
GoatGuy is online now GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
I'm a little confused by this comment…
Are you saying I shouldn't be using any local feedback around the triode?
Can you explain further?
Local negative feedback is a fine thing; it is quite commonly used to "tame" the nonlinearities of the triode (and to some degree, pentodes, too.)

BUT … if you were asking my opinion as to the better of many possible local negative feedback topologies, I prefer the non-bypassed cathode resistor method, and balancing the VA operating point to somewhere between ½ and ⅔ of B+ nominally.

The unbypassed cathode resistor triode stage has a significant drop in gain (which sounds per your discussion as beneficial!) compared to either bypassed resistor biasing or LED (or similar) biasing. Because it is also negative feedback on the most local level, it also works well to linearize the triode gain stage. Try it: you need only temporarily disconnect the CK cathode bypass cap from the 560 Ω or 600 Ω cathode resistor. Remember: output will be lower. But listen for the "nicer" acoustic signature of the triode.
Initially I ran the triode open loop (and so the pentode…), after degenerating the pentode because gain was rather high, then I went back to the triode and degenerated that too.
See above.
IIRC at either bias point, the triode still has headroom, Well after the pentode grid is driven far enough to clip the output.
One thing to consider is … that you really don't want the output of the “front end” to be higher than the negative bias of the output pentode. If VK is running 30 volts above ground (say), then you definitely don't want the output of the prior stage to have more than a ±15 volt swing. Above that, and you get into source material driving the pentode grid into positive-relative-to-cathode territory, which of course causes it to draw real grid current, … which in turn seriously changes the load to the prior driver stage. All of which tends to create both linear and nonlinear distortion.

Hope that helps.
GoatGuy
__________________
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink
  Reply With Quote
Old 12th February 2019, 07:38 PM   #12
mondogenerator is offline mondogenerator  England
diyAudio Member
 
mondogenerator's Avatar
 
Join Date: Sep 2009
Location: GB
GoatGuy,

I can't disagree with any of the facts, my opinion isn't formed yet, regarding sound, but yes I probably could use some gain reduction in gain.
The comment about stage loading and interaction is spot on the mark. The triode stage is loaded down to a degree, i cant recall the exact figures but i think gain was somewhat less than what i had expected.

A couple of days back, I'd tried 6C6B triode, with a mu of 20-25. My thinking was there wasn't enough drive current from the higher mu triodes.

As it happens it's not the case at all haha!

What I learned after a few days pottering about, considering if I should try to use a intermediate follower stage or not, was that the pentode stage was happier with higher gain.

The unbypassed cathode on the triode stage is an easy one to try

Which makes me wonder....

Zener regulated screen voltage might be something I try next, I think I have some 47V zener diodes....May be A string of red LEDs for fine adjustment?

Hmm...food for thought. Thanks.
__________________
I must be a child of good times... the scariest thing I've seen done thus far, has been connecting a cathode bypass capacitor reverse polarity...Then rushing to cut the HT!

Last edited by mondogenerator; 12th February 2019 at 07:44 PM.
  Reply With Quote
Old 12th February 2019, 08:34 PM   #13
GoatGuy is online now GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
PS: when you "try" the unbiassed cathode resistor, remember also to unhook the local negative feedback. Its gain reducing won't be needed (or desired).

PS: In general I tend toward 3-stage small amplifiers such as you are building. One of the more significant reasons (rarely cited!) is that it allows the tube curve-shape asymmetries to be somewhat balanced between signal phases. Better, not best.

The "preamp" stage should be spec'd with "one of the most linear, least power-law" triodes. The next stage should be low-mu, and only "barely amplify at all" (maybe 3× or so) This sets things up for the final stage, well.

GoatGuy

(PS: a zener regulated screen isn't a bad idea, all in all. I've used it quite successfully. Tho design usually calls for a fairly high screen, I tend toward 150 V… or 3 of your Zeners in a string. Remember to feed with a "low enough" resistor.

If your B+ is 300 V, and we're shooting for a 150 V drop, and recognizing that the screen can easily sink 5 ma, then …
E = I R
R = E / I
R = 150 ÷ 0.005
R = 30 kΩ (nearest is 27 kΩ)
Also… tho it is a bit of a stretch, you could use the same trick for the VK of the final valve. Purchase a bunch of 5 volt Zeners, string in series, making your VK = 30 volts. No capacitor and resistor needed for biasing.

Just saying,
GoatGuy
__________________
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink
  Reply With Quote
Old 12th February 2019, 11:42 PM   #14
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
SEP using 6P30B and 5744WB/6C7B
The degeneration provided by that 1M resistor around the driver stage is negligible which is a good thing otherwise the input impedance would be pretty close to Rin which currently it isn't. The open loop gain of that triode is far less than the ratio of Rin/Rfb so it's not doing much of anything except perhaps providing a path for bias current.

What drove the choice of the 5744? I've not found it to be terribly linear tbh..
__________________
"To argue with a person who has renounced the use of reason is like administering medicine to the dead." - Thomas Paine
  Reply With Quote
Old 13th February 2019, 01:10 AM   #15
GoatGuy is online now GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
What drove the choice of the 5744? I've not found it to be terribly linear tbh..
LOL… there's no accounting for why designers (you and I and many of the readers) do what we do. Opportunity is part of it. I would be surprised if you actually went to the trouble of sourcing the almost rare 5744 … it was a WW2 / military tube that was supposed to be miniature, tough-as-nails, direct-circuit-solderable, and almost infinitely EMP/radiation proof. Pretty good characteristics when germanium transistors would go bad just by glancing at them the wrong way. In the height of the megaton cold war era. Just saying…

OK, so the 1 megger does near nil.

Try out the no-cap RK bias on stage 1.
Tho' most commenters might demur, instead of LNFB
(local negative feedback)
to tame the output gain, use the LNFB degenerate cathode bias instead.

Yes, you lose some power.
Yes, there are more cool ways of doing it.

Still … it works pretty well.

For “trial 2”, instead, drop in an appropriate chain-of-Zeners for the RK of the final stage; I've personally found that the sound is somehow "sweeter" for most music program material. Can't quite put my finger on why, but it is nice.

Best of luck to you.
I'm on a flight in the AM…
To someplace 155° West of where I am today.
So… replying isn't much of an option.

Looking forward to recover this thread in 3 weeks!!!

GoatGuy
__________________
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink
  Reply With Quote
Old 13th February 2019, 05:44 AM   #16
mondogenerator is offline mondogenerator  England
diyAudio Member
 
mondogenerator's Avatar
 
Join Date: Sep 2009
Location: GB
Kevinkr,
thanks for you reply. I am not happy with the feedback arrangement as it stands, and considered that as it stands, perhaps it isn't doing much of anything (I.e. i thought 1Meg....not much is getting through that)

GoatGuy,

Firstly, enjoy your holiday!

I may try the zener string for screen voltage, but presently with around 100uF on it, it seems fairly happy.

Since I have B+ of 300V and Vg2 of 250V and Rg2 is 47k, I figure:

(300-250)/47k = ~1.1mA Screen current

(I'm not sure this is the right way to calculate screen current, but its the only wat i can think of at 6:30am.

It sounds like I need to try without cathode bypass firstly.

As for the choice of 5744 tubes...well...I happened across a boatload for a good price and bought them. I'll probably die with the tubes still in the attic, but I've wasted larger sums before and lived...
(I have 6021, 6111, 6112, 6C6B, 6C7B, 6N16B, 6N17B, 6N21B and some rod pentodes, but the 5744 are the vast majority if my tube stock - better tubes such as 6021 or 6N17B may end up being preferred choice for Hifi. For guitar the 5744 might find a good home)
__________________
I must be a child of good times... the scariest thing I've seen done thus far, has been connecting a cathode bypass capacitor reverse polarity...Then rushing to cut the HT!

Last edited by mondogenerator; 13th February 2019 at 05:47 AM.
  Reply With Quote
Old 14th February 2019, 06:58 PM   #17
mondogenerator is offline mondogenerator  England
diyAudio Member
 
mondogenerator's Avatar
 
Join Date: Sep 2009
Location: GB
Hi,

I haven't had much time since last post, but I tried unbypassed cathode, and the reduction in gain is clear. Rk at 820R and I just about make 1.7V. Gain is probably 15 V/V.

Time for the LED. Better performance immediately, but back to a slight excess gain. Catch 22

Then I read the thread through again and This popped out and really made sense.

Quote:
Originally Posted by GoatGuy View Post

PS: In general I tend toward 3-stage small amplifiers such as you are building. One of the more significant reasons (rarely cited!) is that it allows the tube curve-shape asymmetries to be somewhat balanced between signal phases.

The "preamp" stage should be spec'd with "one of the most linear, least power-law" triodes. The next stage should be low-mu, and only "barely amplify at all" (maybe 3× or so) This sets things up for the final stage, well.

Just saying,
GoatGuy
Suffice to say I think I will abandon the first stage, keep the pentode stage and work backwards.

Now that, (surprisingly) I seem to have the pentode working nicely, and I know I need 30V pk-pk to drive it to full output, I'll leave that well alone.
__________________
I must be a child of good times... the scariest thing I've seen done thus far, has been connecting a cathode bypass capacitor reverse polarity...Then rushing to cut the HT!
  Reply With Quote
Old 1st March 2019, 12:07 PM   #18
mondogenerator is offline mondogenerator  England
diyAudio Member
 
mondogenerator's Avatar
 
Join Date: Sep 2009
Location: GB
Default A little update

Hi,

I've had a small amount of tinker time the last week, and I have had a small amount of success in improving the design.

I have now committed to using 6S6BV triode in V1 position, and I have added a divider to drop the B+ from 300V to 240V. This feeds the triode stage. The pentode output stage takes the main 300V B+ supply.
Screen Voltages are supplied via another divider and bypassed to ground, Vg2 set at 245V as before.

Next I split the gain over two triode stages, so that I could include a volume pot between the stages (schematics being drawn over the weekend, and I'll post after they're finished)
In the end i achieve results similar to the two stage design, no improvement or large change.

In the meantime, I have received some better quality OPTs(Thanks Jogi)

Up until now I have been using radio OPTs (which I guess are either wound for 3 Ohm or 15 Ohm loudspeakers)

I tested the radio OPT using dummy loudspeaker load and measuring primary and secondary volts. The radio OPT comes it to 4k2 reflected impedance.(i have several others which work out to 7k, 10k and 14k, all _work_ but I found the 4k and 7k work best.

So with this in mind, I bought some 4k/5k OPTs and hooked them up today....

Well the results were initially not great, so I tried the 5k tap. Small improvement. Still not great. I see soft clipping on the positive half cycle so I assume that grid bias is close to 0V. It seems that revising doesn't help.

So I decide to use the 4 Ohm tap instead....

Great improvement! Back to where I was with small radio OPTs but with a much better frequency response (new OPTs are about 4 times the size)

But I couldn't understand why the new OPTs were so much different.

So I measured Inductance of each OPT with dummy load on the corresponding taps.

Radio OPT: 5.9H (8R load)
Jan OPT: 6H (4k:4R 8R load), 12H with 8R on 8 Ohm tap....
__________________
I must be a child of good times... the scariest thing I've seen done thus far, has been connecting a cathode bypass capacitor reverse polarity...Then rushing to cut the HT!
  Reply With Quote
Old 1st March 2019, 12:11 PM   #19
mondogenerator is offline mondogenerator  England
diyAudio Member
 
mondogenerator's Avatar
 
Join Date: Sep 2009
Location: GB
Now this was somewhat of a surprise, as I had assumed the radio OPTs would have a lack of inductance and a larger L would if anything, be an improvement (I'm sure I've read something that made me believe this to be true)

However, I found that the tubes worked similarly to previous iterations, only by using twice the primary Z I expected to need.

I'm still trying to fathom the exact reason why (any idea?)
__________________
I must be a child of good times... the scariest thing I've seen done thus far, has been connecting a cathode bypass capacitor reverse polarity...Then rushing to cut the HT!
  Reply With Quote
Old 7th March 2019, 03:13 PM   #20
mondogenerator is offline mondogenerator  England
diyAudio Member
 
mondogenerator's Avatar
 
Join Date: Sep 2009
Location: GB
Default Progress? Perhaps.... well an update at least

Soooo.... I figure the inductance may not be the issue, but perhaps OPT DCR instead.

The shiny bright Jan Wuestern ATRA0288 OPTs I bought recently have half the DCR of the cr@ppy radio OPTs I've been using up until now.

Even on a small 2" wideband driver (Visaton FRS5X) the bass warmth comes through

I've had so much fun listening to music I wouldn't necessarily normally listen to that I haven't made huge leaps of progress. (e.g. Dire Straits, Fleetwood Mac, tracks I recall from childhood)

So far I can say I haven't homed in on what some people call valve sound, merely I have enjoyed the relatively clean playback of audio. No-one likes to listen to sine waves, at least without some envelope, and even then, it probably best under 100Hz.

So I have had a little rethink, and opted to add another VAS again using 6S6B-V tubes with their lower mu and better current capability.

The reasoning behind this was that I wanted to incorporate a volume control potentiometer, and I thought that best placement could be between the two front end stages.

SO now I have the inputs stage running Open Loop (about 8 V/V), feeding a pot, to the next stage with LNFB to give a gain of roughly 2V/V, which then drives the Pentode output.

schematic and scope shots attached

The scope shots show waveshape at various points in the circuit as follows:

Ch1 (Yellow) input from Signal Generator
Ch2 (Magenta) Signal at output of 1st stage after DC blocking Capacitor
Ch3 (Blue) Signal at the output of the 2nd stage
Ch4 (Green) Signal at the Pentode anode (AC coupled)

Three input Signal Levels are shown: 1.4V p-p, 2V p-p and ~2.3V p-p (corrsponding to 0.5Vrms, 0.707Vrms and 0.85Vrms)
Attached Images
File Type: gif RISX0084.gif (45.3 KB, 51 views)
File Type: gif RISX0085.gif (41.0 KB, 51 views)
File Type: gif RISX0086.gif (41.8 KB, 52 views)
Attached Files
File Type: pdf 6S6B-V_6P30B_SEP_002_sch.pdf (25.7 KB, 9 views)
__________________
I must be a child of good times... the scariest thing I've seen done thus far, has been connecting a cathode bypass capacitor reverse polarity...Then rushing to cut the HT!
  Reply With Quote

Reply


SEP using 6P30B and 5744WB/6C7BHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Looking for 6P30B Amp Schematic Kevin Mitchell Tubes / Valves 11 17th January 2018 02:09 PM
Anyone have experience with transformers for a 6p30b-r PP Amp? usbekhistan Instruments and Amps 6 12th October 2012 05:49 AM
6P30B-V subminiature tube orson56 Tubes / Valves 0 27th June 2011 08:34 PM


New To Site? Need Help?

All times are GMT. The time now is 11:57 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki