E810F and general pentodes design question - 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 29th December 2013, 06:53 PM   #1
diyAudio Member
 
Joshua_G's Avatar
 
Join Date: Nov 2008
Location: Small village, Israel
Default E810F and general pentodes design question

Right now I'm trying to design the first stage of an MM phono stage with 600R LCR RIAA compensation network. Alas, my knowledge about pentodes circuits design is wanting.

So here are few questions.

1. On Philips E810F datasheet (the only one I have for this tube), under operating characteristics, it mentions having +12.5V on the control greed. What is this voltage for? Does it really necessary, or recommended, for a phono first stage? (I've seen some schematics of this tube in a phono first stage, some schematics apply positive voltage to the greed, while other schematics don't).

2. If my memory serves me right, I remember reading somewhere that applying stabilized voltage to greed #2 reduces harmonic distortion. Is it correct? If yes, is it meaningful at such low levels?

3. Should I choose stabilized voltage for greed #2, one option is to use VR tube. VR tubes are somewhat noisy, but it's easy to filter out this noise with a capacitor parallel to the tube. What I don't know is if those tubes radiate noise to the surroundings. If this is the case, I'll definitely not use any VR tube inside a phono stage enclosure.

4. I have no clue how to locate the most linear operating point for pentode (if there is such point). To ask that question differently, according to what is it best to choose the operating point (anode voltage, greed #2 voltage and bias) of a pentode?

5. Is it necessary, or is it recommended, that greed #2 voltage will be higher than the anode voltage?

6. When the output of that first stage is loaded by a resistive (or impedance) load which is significantly lower than the anode loading resistor (via a coupling capacitor), it appears to me that the output signal of the tube will be attenuated by the ratio of the anode loading resistor to the load. Is that correct? If not, how do I calculate the attenuation of a given load?
  Reply With Quote
Old 29th December 2013, 07:39 PM   #2
diyAudio Member
 
Join Date: Apr 2004
Location: Monroe Township, NJ
First, have you considered the S/N issue pentode partition noise raises?

Use the plate curves to establish a rational operating point. Regulating g2 B+ at a fraction of anode B+ yields the best open loop linearity.

A 600 Ω LCR RIAA module must be associated with a 600 Ω impedance at either the I/P or the O/P, to function properly. Best performance will be obtained with 600 Ω at the I/P. How do you propose to fulfill the requirement?
__________________
Eli D.
  Reply With Quote
Old 29th December 2013, 07:46 PM   #3
diyAudio Member
 
the_manta's Avatar
 
Join Date: Nov 2006
Location: Munich, Bavaria
1. There may be a positive voltage, but the effective potential between the grid and the cathode is negative. That's because the positive Voltage is "annihilated" with a Cathode resistor. It's all in the datasheet, you just have to look
Cathode resistor: 360R
Anode current: 35mA
Grid #2 current: 5mA
Thus, the voltage at the cathode: 360*(35+5) = 14,4V
This means: 12.5 - 14,4 = -1.9V effective Grid potential.

Why you do that ? The Cathode resistor provides current feedback (series derived - series applied). This stabilizes (and lowers) the transconductance and increases the input and output resistance.

2. Kinda. If the voltage @ Grid 2 is distorted, than you will get distortion. You can also the Grid 2 to lower the distortion (feedback as in Blumlein's UL). The Problem is the very unlinear current which the G2 draws at lower anode voltages. So - Yes a regulated supply is supportive.

3. Do NOT use a parallel cap considering VR tubes. You will build an oscillator. The max. value for a cap is described in the datasheet. Just a small value is allowed.
I would leave out any gas stabilizers in a phono pre. People may have different opinions here but it just makes everything more difficult. MOSFETs are so damn cheap -> just built a Maida regulator or even cheaper with TL431 (http://www.changpuak.ch/electronics/...ads/QO-408.pdf)

4. Just look at the diagrams in the datasheet. I'd suggest you look around the 35mA range.

5. Grid #2 Voltage should be lower than anode voltage.

6. For a pentode, the amplification is roughly Vu = gm*Rout

What about that ?:

Click the image to open in full size.
__________________
Терпенье и труд все перетрут
  Reply With Quote
Old 29th December 2013, 08:07 PM   #4
diyAudio Member
 
Joshua_G's Avatar
 
Join Date: Nov 2008
Location: Small village, Israel
Hi Eli,
Thank you.

Quote:
Originally Posted by Eli Duttman View Post
First, have you considered the S/N issue pentode partition noise raises?
Yes, I did.

Quote:
Originally Posted by Eli Duttman View Post
Use the plate curves to establish a rational operating point. Regulating g2 B+ at a fraction of anode B+ yields the best open loop linearity.
Thank you.

Quote:
Originally Posted by Eli Duttman View Post
A 600 Ω LCR RIAA module must be associated with a 600 Ω impedance at either the I/P or the O/P, to function properly. Best performance will be obtained with 600 Ω at the I/P. How do you propose to fulfill the requirement?
Thank you, I know it must be loaded with 600 Ω at one side.
I'm considering 3 options:
1. Large anode loading resistor -> coupling capacitor -> 600 Ω LCR module -> 600 Ω resistor.
2. 600 Ω anode loading resistor -> direct coupled to 600 Ω LCR module (with high voltage capacitors inside it).
3. Step-down transformer.

Which option will give me a larger voltage gain at the output of the LCR module?
  Reply With Quote
Old 29th December 2013, 08:18 PM   #5
diyAudio Member
 
Joshua_G's Avatar
 
Join Date: Nov 2008
Location: Small village, Israel
Hi there,
Thank you.

Quote:
Originally Posted by the_manta View Post
1. There may be a positive voltage, but the effective potential between the grid and the cathode is negative. That's because the positive Voltage is "annihilated" with a Cathode resistor. It's all in the datasheet, you just have to look
Cathode resistor: 360R
Anode current: 35mA
Grid #2 current: 5mA
Thus, the voltage at the cathode: 360*(35+5) = 14,4V
This means: 12.5 - 14,4 = -1.9V effective Grid potential.

Why you do that ? The Cathode resistor provides current feedback (series derived - series applied). This stabilizes (and lowers) the transconductance and increases the input and output resistance.
What stabilizes (and lowers) the transconductance and increases the input and output resistance the cathode bias resistor?
How does applying positive voltage to the control greed counteracts it?

Quote:
Originally Posted by the_manta View Post
2. Kinda. If the voltage @ Grid 2 is distorted, than you will get distortion. You can also the Grid 2 to lower the distortion (feedback as in Blumlein's UL). The Problem is the very unlinear current which the G2 draws at lower anode voltages. So - Yes a regulated supply is supportive.

3. Do NOT use a parallel cap considering VR tubes. You will build an oscillator. The max. value for a cap is described in the datasheet. Just a small value is allowed.
I would leave out any gas stabilizers in a phono pre. People may have different opinions here but it just makes everything more difficult. MOSFETs are so damn cheap -> just built a Maida regulator or even cheaper with TL431 (http://www.changpuak.ch/electronics/...ads/QO-408.pdf)
Thank you.

Quote:
Originally Posted by the_manta View Post
4. Just look at the diagrams in the datasheet. I'd suggest you look around the 35mA range.
What to look for on the anode curves? (I know how to look at triodes anode curves but I have no clue how to do it on pentodes).

Quote:
Originally Posted by the_manta View Post
5. Grid #2 Voltage should be lower than anode voltage.

6. For a pentode, the amplification is roughly Vu = gm*Rout
Thank you.

Quote:
Originally Posted by the_manta View Post
What about that ?:

Click the image to open in full size.
I'm glad you brought up this schematic. I know it for a while. One thing bugs me about it: I cannot figure out how the bias works there.
  Reply With Quote
Old 29th December 2013, 08:33 PM   #6
diyAudio Member
 
Joshua_G's Avatar
 
Join Date: Nov 2008
Location: Small village, Israel
Quote:
Originally Posted by the_manta View Post
6. For a pentode, the amplification is roughly Vu = gm*Rout
What is Rout in this case? Is it the anode loading resistor parallel to the load resistor?
  Reply With Quote
Old 29th December 2013, 08:38 PM   #7
diyAudio Member
 
Miles Prower's Avatar
 
Join Date: May 2005
Location: USA
Blog Entries: 6
Quote:
Originally Posted by Joshua_G View Post
1. On Philips E810F datasheet (the only one I have for this tube), under operating characteristics, it mentions having +12.5V on the control greed. What is this voltage for? Does it really necessary, or recommended, for a phono first stage? (I've seen some schematics of this tube in a phono first stage, some schematics apply positive voltage to the greed, while other schematics don't).
They wanted to lengthen the "tail" for some reason.

VK= 360(35E-3 + 5E-3)= 14.4V
VGK= 12.5 - 14.4= -1.9V

1.9/40E-3= 47R5

They wanted to avoid a 47R5 bias resistor. Maybe since it's easier to bypass a 360R resistance?

Quote:
2. If my memory serves me right, I remember reading somewhere that applying stabilized voltage to greed #2 reduces harmonic distortion. Is it correct? If yes, is it meaningful at such low levels?
That's right. If V2K isn't stabilized, you get a sort of remote cutoff characteristic with its greater variations in gm that degrade sonic performance. Whether it makes a difference in a small signal application depends. At very low signal levels, then it's not so much of a problem. A stiff voltage divider will usually serve the purpose of stabilizing V22. Large signal pentode stages are the ones where you'll need active regulation.

Quote:
3. Should I choose stabilized voltage for greed #2, one option is to use VR tube. VR tubes are somewhat noisy, but it's easy to filter out this noise with a capacitor parallel to the tube. What I don't know is if those tubes radiate noise to the surroundings. If this is the case, I'll definitely not use any VR tube inside a phono stage enclosure.
For a very low level stage like this one, it's not necessary, and not desirable from a noise PoV. Paralleling a VR tube with a capacitor, with its ballast resistor, makes an oscillator.

Quote:
4. I have no clue how to locate the most linear operating point for pentode (if there is such point). To ask that question differently, according to what is it best to choose the operating point (anode voltage, greed #2 voltage and bias) of a pentode?
Loadlines, loadlines, loadlines: that's how hollow state design is always done. (See attached: 6CB6 loadline)

The plate xfer characteristics are usually more useful for designing audio gain stages, since the plate characteristics are frequently provided for entirely different purposes. The type you mention here says right there in its spec sheet it was designed for use as a wideband amp, not an audio amp. As such, it was intended to pull some stiff plate currents so that it could drive load capacitance, and produce useful voltage gain from smallish plate loads that improve high frequency performance.

For low distortion, you need to load the plate lightly (minimize delta-Ip) and get that screen voltage down. If you're doing pentode designs, expect that there will be a lot of empirical adjustments along the way to finalizing a design -- moreso than for doing triode based designs.

Quote:
5. Is it necessary, or is it recommended, that greed #2 voltage will be higher than the anode voltage?
Not strictly necessary, but the more time the plate voltage stays above the screen voltage the better.

Quote:
6. When the output of that first stage is loaded by a resistive (or impedance) load which is significantly lower than the anode loading resistor (via a coupling capacitor), it appears to me that the output signal of the tube will be attenuated by the ratio of the anode loading resistor to the load. Is that correct? If not, how do I calculate the attenuation of a given load?
Any load that the stage drives will appear in parallel with the plate load resistor. The parallel load becomes the effective plate load, and the heavier that load, the lower the stage gain:

AV= (RP(eff) || rp) X gm

I would question the suitability of a pentode for MM phono duty. Pents have an additional source of noise called partition noise in addition to the sources that plague triodes. It usually doesn't make much difference at RF, since the noise decreases with increasing frequency, but for sensitive RF designs, triodes are preferred.

You could also consider the cascode: all the gain, and much less of the CMiller of a pentode, but none of the partition noise.
Attached Images
File Type: png 6CB6-Plate-Xfer-68K.png (178.4 KB, 311 views)
__________________
There are no foxes in atheistholes
www.dolphin-hsl.com
  Reply With Quote
Old 29th December 2013, 09:04 PM   #8
diyAudio Member
 
the_manta's Avatar
 
Join Date: Nov 2006
Location: Munich, Bavaria
Quote:
Originally Posted by Joshua_G View Post
What is Rout in this case? Is it the anode loading resistor parallel to the load resistor?
Exactly. It is the sum of all the loads between cathode and anode. If there are capacitors/inductors you have to take account of different loads at different frequencies, too
__________________
Терпенье и труд все перетрут
  Reply With Quote
Old 29th December 2013, 09:11 PM   #9
diyAudio Member
 
Joshua_G's Avatar
 
Join Date: Nov 2008
Location: Small village, Israel
Hi Miles,
Thank you.

Quote:
Originally Posted by Miles Prower View Post
They wanted to lengthen the "tail" for some reason.

VK= 360(35E-3 + 5E-3)= 14.4V
VGK= 12.5 - 14.4= -1.9V

1.9/40E-3= 47R5

They wanted to avoid a 47R5 bias resistor. Maybe since it's easier to bypass a 360R resistance?
Sorry, I don't get your reply. When a positive voltage is applied to the control grid, than a larger cathode resistor is needed, which is more difficult to bypass.

Quote:
Originally Posted by Miles Prower View Post

For low distortion, you need to load the plate lightly (minimize delta-Ip) and get that screen voltage down.
How much lightly? 100 Ω? 1KΩ?
On top of that, the lower load, the lower the gain and I do need substantial gain in this application.

Quote:
Originally Posted by Miles Prower View Post

Not strictly necessary, but the more time the plate voltage stays above the screen voltage the better.
How much down? 10V on g2? 20V? 50V?

Quote:
Originally Posted by Miles Prower View Post
I would question the suitability of a pentode for MM phono duty. Pents have an additional source of noise called partition noise in addition to the sources that plague triodes. It usually doesn't make much difference at RF, since the noise decreases with increasing frequency, but for sensitive RF designs, triodes are preferred.
Thank you.
That pentodes are less suitable for a phono stage is the prevailing thinking. Yet, both Thorsten Loesch and Thomas Mayer (Vinylsaviour) used pentodes in that application, so I wonder.
  Reply With Quote
Old 29th December 2013, 09:14 PM   #10
diyAudio Member
 
Joshua_G's Avatar
 
Join Date: Nov 2008
Location: Small village, Israel
Quote:
Originally Posted by the_manta View Post
Exactly. It is the sum of all the loads between cathode and anode. If there are capacitors/inductors you have to take account of different loads at different frequencies, too
Thank you.
Indeed, this is how LCR module works it presents different load at different frequencies.
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
Sub Enclosure General Design Shape jshack Construction Tips 2 13th February 2012 02:38 AM
John Broskie's headphone design using pentodes Brit01 Headphone Systems 14 15th July 2009 02:28 AM
Beam pentodes and suppressor-grid pentodes of same type jon_010101 Tubes / Valves 3 2nd May 2009 11:18 PM
Design using EL42 output pentodes phil_2 Tubes / Valves 61 13th November 2008 11:10 AM
Fostex, ATC, Seas... General design question - Please advise. amo Multi-Way 5 15th July 2003 12:53 AM


New To Site? Need Help?

All times are GMT. The time now is 04:50 PM.


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