Pentode Plate Load - 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 4th June 2007, 11:45 PM   #1
diyAudio Member
 
dsavitsk's Avatar
 
Join Date: Jan 2005
Location: Hartford
Default Pentode Plate Load

I know most of the basics of working with triodes, and triode strapped pentodes, but how does one go about picking the plate load for a pentode (assuming a grounded cathode stage) be it a resistor, transformer, or choke?
  Reply With Quote
Old 5th June 2007, 12:17 AM   #2
diyAudio Member
 
the_manta's Avatar
 
Join Date: Nov 2006
Location: Munich, Bavaria
Hi !
You do that in a similiar way as for triodes. You pick your curves and draw a suitable loadline and pick a symmetrical operating point.
The abilty of a pentode's anode to control the current is very low, resulting in an extremely low punch-through effect and an accordingly very high . Result is that the amplification is mostly depending on the Load on the grids ability to change the current --> gm. Short equation: A = S*Ra
Screen grid voltage must be fixed to cathode n a kind of way. Either a cap from screen to ground or a gas stabilizing tube does this job perfectly.
A good pentode working in inductive load is one of the best stages ever. With a high gm tube and larger currents you are able to drive evrything. A C3M can drive a 300B without problems and with very low distortion.
I always prefer pentodes. It's the more decent, natural und dynamic sound. Not that warm, plentyful triode sound with the overemphasized midrange and voices.

Regards, Simon
__________________
Терпенье и труд все перетрут
  Reply With Quote
Old 5th June 2007, 08:41 PM   #3
diyAudio Member
 
pmillett's Avatar
 
Join Date: May 2005
Location: Dallas (but I am not a Texan!)
To expand a little bit on Simon's reply... there is maybe an additional degree of freedom as compared with a triode, which has more-or-less fixed voltage gain.

You can pick the plate load based on desired gain (high = high gain, low = low gain), or on desired Zout (low=low, since Rp is typically very high, Zo is dominated by Rl). Of course you also have to consider the B+ and plate current as well.

For example, for a driver stage where you want low output impedance, you can pick the lowest resistance that gets you a reasonable voltage swing.

Or, for a mic preamp, you might want the highest load you can do within B+ limitations to get really high gain.

Yet another choice is to optimize for lowest distortion, by eyeballing the curves. Some pentodes can be amazingly linear operated over the right spot.

Of course, usually you wind up doing a compromize between all of these things.

Pete
  Reply With Quote
Old 5th June 2007, 09:00 PM   #4
diyAudio Member
 
dsavitsk's Avatar
 
Join Date: Jan 2005
Location: Hartford
Thanks, that's very helpful. Additionally, I seem to see the screen connected to B+ through a largish resistor (50K on up) and bypassed to ground through a smallish cap (100n or so.) Any tips on picking those, or other ways to connect things up? And, what exactly is that resistor doing. Does it keep the screen at a high voltage (with no current flowing), or does it drop lots of volts with a small amount of current flowing, and iof so, can it be connected to ground instead.

My current application is a guitar amp, so distortion is not a bad thing, I don't think.

Quote:
Originally posted by pmillett
since Rp is typically very high, Zo is dominated by Rl). Of course you also have to consider the B+ and plate current as well.
Rl is the load?
  Reply With Quote
Old 5th June 2007, 09:19 PM   #5
EC8010 is offline EC8010  United Kingdom
diyAudio Moderator Emeritus
 
EC8010's Avatar
 
Join Date: Jan 2003
Location: Near London. UK
Yes, RL is the load.
__________________
The loudspeaker: The only commercial Hi-Fi item where a disproportionate part of the budget isn't spent on the box. And the one where it would make a difference...
  Reply With Quote
Old 5th June 2007, 10:31 PM   #6
diyAudio Moderator Emeritus
 
ray_moth's Avatar
 
Join Date: Jan 2004
Location: Jakarta
Quote:
I seem to see the screen connected to B+ through a largish resistor (50K on up) and bypassed to ground through a smallish cap (100n or so.) Any tips on picking those, or other ways to connect things up? And, what exactly is that resistor doing.
With small signal pentodes, the screen current is usually about 20% of the plate current. A typical plate load for a pentode might be 100k. The screen resistor might be in the region of 470k to 680k, to provide a screen voltage that is in the region of the plate voltage but somewhat lower.

Also, the screen needs to be AC-connected (decoupled) to the cathode. With a fully-bypassed cathode bias resistor, the screen can be decoupled through a capacitor to ground; however, if some or all of the cathode resistor is not bypassed (e.g. where global NFB is is injected into the cathode), the screen needs to be decoupled to the cathode itself. A suitable sized cap for this purpose would be ~0.1uF. The quality of this cap is important and its value can have a strong influence on low frequency stability if the pentode is inside an NFB loop. It's one more thing you may have to fiddle with.
  Reply With Quote
Old 6th June 2007, 03:00 AM   #7
diyAudio Member
 
Miles Prower's Avatar
 
Join Date: May 2005
Location: USA
Blog Entries: 6
Quote:
Originally posted by dsavitsk
Thanks, that's very helpful. Additionally, I seem to see the screen connected to B+ through a largish resistor (50K on up) and bypassed to ground through a smallish cap (100n or so.) Any tips on picking those, or other ways to connect things up?
If you have the characteristic curves, you can determine what the screen current will be, and so you simply use a series resistor to drop the Vpp to Vsgsg. If using a fully bypassed cathode bias resistor, my own preference is to stiffen the screen voltage with a voltage divider, especially if there's gonna be a significant plate current swing. A series screen resistor'll convert any pentode into a (more or less) remote-cutoff-type device. Not a good thing for low distortion. Click the image to open in full size.

(Unless that's what you need, as in the IF strip of a receiver.)

If the cathode isn't bypassed (pentode cathode follower, as in the upper half of a Kimmel) then screen voltage swing is essential if you want to prevent a type of ultralinear operation that will reduce gain and raise Zo. In that case, a series resistor + healthy screen coupling to the cathode is required.

The screen impedance is pretty up there, so a big capacitor isn't required. Guesstimate it as: Rsg= (Vsgk/Isgq || Rsg), and scale the capacitor according to your low frequency requirements.

There are other cases where you might have to tailor the screen bypass for other considerations since the screen is kind of a control grid, and the RC combination can insert an extra pole into the transfer characteristic of the amp, with sometimes unexpected results.

Quote:
My current application is a guitar amp, so distortion is not a bad thing, I don't think.
Go with the simple series dropping resistor and see what happens. If it sounds too gawdawful, then try stiffening up that screen voltage.
__________________
There are no foxes in atheistholes
www.dolphin-hsl.com
  Reply With Quote
Old 6th June 2007, 04:53 PM   #8
cerrem is offline cerrem  United States
diyAudio Member
 
Join Date: Jun 2005
Location: San Diego, CA
One thing to keep in mind with pentodes....You want to stay in the SATURATION region for the entire load-line swing....
The SATURATION region is the part of the IV curve that goes horizontal after the knee or close to that....It has a small slope...that slope is the plate resistance and it is quite high...An ideal pentode would have totally flat horizontal curves indicating a perfect current source of infinite impedence...
Well the key curve to look at is your 0v bias curve...since this is the end of the line for your load-line swing.....assuming your staying away from driving for grid current...
YOu want to maximize your AC swing for the must current and most voltage..so when the load-line approches the 0v bias curve cross it to the right of the knee and make sure your still in Saturation.... if your plate load number is too big you will interect you 0v bias curve when it is still in the TRIODE region when it is almost vertical....this will not only introduce more non-linearity, but the plate resistance of the pentode will drop low and cuase more issues...

Chris
  Reply With Quote
Old 6th June 2007, 09:57 PM   #9
diyAudio Member
 
Miles Prower's Avatar
 
Join Date: May 2005
Location: USA
Blog Entries: 6
Quote:
Originally posted by cerrem
if your plate load number is too big you will interect you 0v bias curve when it is still in the TRIODE region when it is almost vertical....this will not only introduce more non-linearity, but the plate resistance of the pentode will drop low and cuase more issues...

Chris
(See attached)

This may or may not be a problem. If doing something like a DC voltage amp for a regulator, then dunworryboudit: the Vgk will always stay very close to the nominal bias value. That's how the thing works: it always tries to move the voltage to maintain that particular value of voltage differential.

If you can't be certain that some idgit won't connect to some source that outputs more than 2.5Vp, then you can either do one of two things: draw a loadline that goes through Vgk= 0 while still within saturation. Here, you wouldn't want to do that because we've already got one helluvagood loadline there. (The one in Blue, not the Green one.)

Now, it would be nice if we could replace the Vgk= -1.0Vdc line with the Vgk= 0V line. Since this particular characteristic was made against Vsgk= 150Vdc, we can do just that. If we want to drop the whole thing by one volt, pick off the loadline Ip at a Vgk= -4.0V. Then go to the "Average Transfer Characteristic" chart that plots Vgk against Ip at various screen voltages. Find out where the -3.0V line intercepts the plate current for Vgk= -4.0V @ Vsgk= 150V. By means of interpolation, this comes quite close to a Vsgk= 120Vdc (actually something like 119. something, something, something but don't worry about that since VTs aren't all that accurate, and VT circuits tend to be rather "forgiving").

Once you have your new screen voltage, next find from the chart the Vgk that'll give the original design Ipq for the new Vgkq bias.

This should preserve the original design performance (or come close enough for tweaking the Vsg for minimal THD if that's what you want) and close enough so that you won't be poofing screens by operating too close to the "triode" region.
Attached Images
File Type: png 6au6-loadline.png (79.8 KB, 689 views)
__________________
There are no foxes in atheistholes
www.dolphin-hsl.com
  Reply With Quote
Old 6th June 2007, 10:42 PM   #10
diyAudio Member
 
jeff mai's Avatar
 
Join Date: Aug 2002
Location: Melbourne
I think of pentodes liking constant voltage, whereas triodes "like" constant current. Pentodes generally have lower distortion (and particularly lower odd harmonics) with smaller plate loads. Of course, with constant voltage you get no output.

I choose the smallest plate load that gives me enough output voltage swing for my application. Lower screen voltages generally produce less distortion, but then there is less input headroom. It's a balancing act. And I recommend trying VR tubes for the screen supplies. They were made for this service and work (and sound) very well.

Unlike what some others have advised I try to stay COMPLETELY away from the knees in the curves. This usually means higher plate voltages and smaller plate resistors (and hence lower gain) than what is recommended as an operating point in the data sheets.

The data sheets try to maximise the gain by aiming the load line at one of the knees. I try to aim the high current end of the loadline as near to the flat part of the 0V curve as possible. If you don't need the input headroom this gives, use a lower screen voltage.

I'm sure some will disagree with the above method, but I say try it. I think you'll like it.
  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
how to connect an EL34 pentode into a triode and 100ohm resistor between plate and g2 jarthel Tubes / Valves 15 26th October 2013 03:44 PM
How do I design a pentode constant current source plate load? Bitrex Tubes / Valves 16 22nd January 2009 09:10 AM
CCS Plate Load Jack Thomas Tubes / Valves 4 21st May 2007 11:39 PM
Low plate voltage pentode (<40v) for diff pair? Rescue Toaster Tubes / Valves 3 2nd September 2006 01:56 AM


New To Site? Need Help?

All times are GMT. The time now is 12:19 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