push pull power amp design question (a lot of actually :) ) - 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 24th September 2008, 09:31 PM   #1
diyAudio Member
 
Join Date: Feb 2008
Question push pull power amp design question (a lot of actually :) )

Hi all,
first of all I want to apologize for my english, but not a native speaker..
I'm trying to design a push pull power amp for my guitar...I started taking an "already working" design (the one I attached here), but I have a lot of question regarding it and how to understand how it works (and so make a sort of reverse enginering of it!).


The power amp should be rated (at most) 50W class AB Push-pull
I'd like to use EL34 tubes and these are the parameters:
Vanode = 480V
Vscreen = 475V
Vbias = -48V (not pretty sure about how to calculate this value...)
OT primary Z=4KOhm

here are my questions...(try to be as clear as I can)
1) How I can calculate the "right" Vbias ? (If I'm not wrong a ball park figure should be Vanode/10)? Which resistor technology to use on cathode? (metal film? carbon film? what else? they should act as a fuse in case of bias failure so they are low wattage, but in case of failure shoud become open circuit melting)
2) What are the R3 and R4 (220K) used for? and what if I change their value?
3) R9 and R10 (5.6 K) act as a low pass filter with the capacitance of the grid? (are some standard values used? or If I want to change tube type (i.e. KT88, or 6L6 provided I have enough current for heater) Do I have to change them?
4) do you think that master volume is ok? (C1 and C2 are at least 10 times bigger than the one on Phase Inverter (missing in the drawing) to prevent at most any frequency loss.
5)How can I calculate the power that this amp can give me with this values? any formula? (El34 tubes, Va = 480V, Vs=475, Vb = -48, OT primary Z = 4Kohm)?
6) What is the parameter on data sheet that points the Bias current? I mean the MAX current drown from the bias circuit by the tube?
7) How can I drow the composite load line for this power amp? The bigger trouble is that reading data sheet (pentode ) I can find Ia/Va curves for some Vg2 values..that are always very different from the one I want to use (in this case 475 V...How can I find or draw the right curves I need to use??)
8) what if I want to put a pentode/triode switch? Which values I have to check to be sure not to be uside tube parameter? (same voltages on Anode and Bias)

Thanks in advance for your help.
Krusty
Attached Files
File Type: zip power_amp1.zip (7.3 KB, 101 views)
  Reply With Quote
Old 24th September 2008, 11:33 PM   #2
DougL is offline DougL  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Wheaton IL.
Blog Entries: 30
I will get the ball rolling.
Quote:
1) How I can calculate the "right" Vbias ? (If I'm not wrong a ball park figure should be Vanode/10)? Which resistor technology to use on cathode? (metal film? carbon film? what else? they should act as a fuse in case of bias failure so they are low wattage, but in case of failure should become open circuit melting)
Bias is Calculated by creating a load line on the tube transfer graph, and choosing an operating point. This is not trivial, but there are posts on the forum to help.
I would choose a non inductive wire wound as a cathode resistor. However, I think almost any resistor will fail open in that position.
Quote:
2) What are the R3 and R4 (220K) used for? and what if I change their value?
I do not know the correct term. The resistors feed bias to the output tubes. Too high a resistor value and the bias drifts due to grid current. Too low, and the bias supply acts as a load on the previous stage, potentially limiting drive voltage and increasing driver distortion. The Tube data sheet (in this case EL34) has a max value listed.
Quote:
3) R9 and R10 (5.6 K) act as a low pass filter with the capacitance of the grid? (are some standard values used? or If I want to change tube type (i.e. KT88, or 6L6 provided I have enough current for heater) Do I have to change them?
R9 and 10 are called Grid stoppers. They are present to prevent high frequency oscillation. The values range from 50R to 10K. R9/10 do limit current in cases where the amp is driven hard. If the grids go positive, current can flow from the tube to the bias.
Quote:
6) What is the parameter on data sheet that points the Bias current? I mean the MAX current drown from the bias circuit by the tube?
During normal operation, the resistance of the grid is greater than 10 times the max value of grid resistor listed in the data sheet.

A note on setting the bias. The Maximum Power for an EL34 is 35W. Power = V (b+) * I (bias current). The 10R resistors are there to measure the bias current. At some level, you just clip a voltmeter across R5 or R6 and adjust the pot until the current is greater than 0 and less that the amount that would exceed 35 watts.

I hope this was some help.

Doug
__________________
Scienta sine ars nihil est - Science without Art is nothing. (Implies the converse as well)
Mater tua criceta fuit, et pater tuo redoluit bacarum sambucus
  Reply With Quote
Old 25th September 2008, 12:16 AM   #3
diyAudio Member
 
Join Date: Oct 2004
Location: Adelaide South Oz
Bias Voltage of about Va/10 is a good estimated value.

EL34 is a 25 Watt (not 35 Watt) rated tube. For guitar amp work they should be run conservatively. What I generally do is work out the 70% maximum dissipation figure, that is, 25 x 0.70 = 17.5 Watts.

Then from Va = 480 Volts calculate idle current from 17.5 / 480 = 36.5 mA. I would probably set for 35mA idle per tube.

I see that you tied the supressor grid (G3) to the bias supply. Many people just connect it to the cathode (0 Volts) BUT connecting it to the negative bias supply is a good idea. I usually go part way between these extremes and connect G3 to about -12 volts.

I would get rid of the "Cross Line" Master Volume control shown on your diagram, there are much better ways to do a Master Volume control (Bootstrapped control on the input to the Phase Splitter is what I prefer).

I would provide individual bias adjustments for each EL34 so that you are not restricted to using matched pairs.

Hope this of some help to you.

Cheers,
Ian
  Reply With Quote
Old 25th September 2008, 01:13 AM   #4
Banned
 
Join Date: Jan 2008
Blog Entries: 2
Learning to design amplifiers starts with learning to ask what an amplifier is for.

By this I mean:

What is the input signal level?

What power output is required into what load?

What is the bandwidth? (A given in audio)

and

What is the source of power?

The output power and the load tell you what voltage swing is required in the output stage, and what the current will be.

The input signal level tells you what gain is required.

The source of power tells you whether additional power conditioning circuitry is required, and influences the choice of topology e.g. push-pull, single ended.

You need to be familiar with conversions between RMS, peak-to-peak and peak nomenclatures and understand how they relate to DC. You need to understand the concept of output impedance, short circuit current and open circuit voltage.

There are 3 primary configurations of amplifiers, in the case of BJTs these are common emitter (emitter follower), common collector and common base. These have their equivalents in valves, fets, MOSFETS, IGBTS etc. which are all 3-terminal devices. They all have radically different characteristics, some desirable, some less so.

Once the characteristics of at least the common collector and emitter follower (or valve equivalents) are absorbed, a foundation for devising an architecture will have been achieved. Central to understanding these configurations is an understanding of how bias levels and operating points are designed and established. This is beyond the scope of a reply in a forum, but there are good straightforward explanations available. The techniques are similar from device type to device type but vary according to the physics of the particular device under consideration, some knowledge of which is generally thought to aid understanding. Valve designs are often complicated by the necessary inclusion of transformers.

Working through an example design from output to input calculating DC and AC currents and voltages will start to give you the kind of insight you need to appreciate an existing design. All the topologies are described in many places complete with the equations required. Once you go through the process of calculating what you need and how you are going to get it you have the tools to back-calculate what you will get from somebody else's circuit. Set yourself some exercises, of get a book.

There are many different ways of using active devices; so many that it is not possible to become an expert overnight.

w
  Reply With Quote
Old 25th September 2008, 08:42 AM   #5
cerrem is offline cerrem  United States
diyAudio Member
 
Join Date: Jun 2005
Location: San Diego, CA
One important thing to keep in mind....
I suspect the voltages you listed are IDLE voltages....
IDLE voltage mean very little when designing amps....
You just need to insure that they are within the safe rattings of the tubes...
You need to design at full power output.... Those voltages will be lower due to the voltage regulation of the supply....
Screen currents will go up tremendously...
For example i know of an EL34 amp running at 460V on the plates at idle and 455V on the screens also at idle...
At full clean power output the screens are roughly 400V and the plates around 430V... The screen currents go up to roughly 24mA per EL34 at full power output...Also keep in mind that if you crank this amp to 10 for full distorion, you will now have square waves...and the screen currents will go way up, because the plate and screen are at same potential, thus acting like a triode and the screen will hog most of the current.....
The bias supply will also dip also.....This is why I prefer biasing the tubes at full power output and seeing what is actually going on durring operation...

Chris
  Reply With Quote
Old 25th September 2008, 05:37 PM   #6
diyAudio Member
 
Join Date: Feb 2008
Quote:
Originally posted by cerrem
One important thing to keep in mind....
I suspect the voltages you listed are IDLE voltages....
IDLE voltage mean very little when designing amps....
You just need to insure that they are within the safe rattings of the tubes...
You need to design at full power output.... Those voltages will be lower due to the voltage regulation of the supply....
Screen currents will go up tremendously...
For example i know of an EL34 amp running at 460V on the plates at idle and 455V on the screens also at idle...
At full clean power output the screens are roughly 400V and the plates around 430V... The screen currents go up to roughly 24mA per EL34 at full power output...Also keep in mind that if you crank this amp to 10 for full distorion, you will now have square waves...and the screen currents will go way up, because the plate and screen are at same potential, thus acting like a triode and the screen will hog most of the current.....
The bias supply will also dip also.....This is why I prefer biasing the tubes at full power output and seeing what is actually going on durring operation...

Chris
Thanks a lot for your help!!
Yes you are right...the values I posted are at IDLE...
But where can I find pentode curves with Vg2 = 475 V ? In order to try to design my load line? and then how can I calculate the values at full power? (to have both load lines (idle and full power?)
Krusty.
  Reply With Quote
Old 25th September 2008, 06:07 PM   #7
cerrem is offline cerrem  United States
diyAudio Member
 
Join Date: Jun 2005
Location: San Diego, CA
Quote:
Originally posted by krusty75


Thanks a lot for your help!!
Yes you are right...the values I posted are at IDLE...
But where can I find pentode curves with Vg2 = 475 V ? In order to try to design my load line? and then how can I calculate the values at full power? (to have both load lines (idle and full power?)
Krusty.

I can assure you your screen voltage will dip mucho lower than that 475..... Also you must account for the 1K screen resisitors, since these will drop the voltage even further.....
EL34 pentode curves are not well published for higher screen voltages..... I can try to post further info if I find it....

Chris
  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
push-pull design nukaidee Multi-Way 31 9th May 2009 10:18 AM
6L6 Push pull -- Power tranny question guitvinny Tubes / Valves 53 7th June 2006 01:34 AM
modify parallel push-pull EL84 to single push-pull chungtat Tubes / Valves 12 4th November 2005 12:25 AM
833 push pull design help sickss Tubes / Valves 19 7th March 2005 04:25 PM
Power rating for push pull configuration question miked15 Tubes / Valves 4 15th May 2002 03:50 AM


New To Site? Need Help?

All times are GMT. The time now is 02:02 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