Bob Cordell's Power amplifier book - Page 239 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 2nd April 2012, 12:15 AM   #2381
diyAudio Member
 
moschfet's Avatar
 
Join Date: Jul 2008
Location: Hannover
Hello Bob,

I bought your book and I'm thrilled. Even as a German I was able to understand the most immediately. The book has saved me a lot of work and explains many things I still wanted to know.

Thanks
__________________
Regards http://xipix.de/
  Reply With Quote
Old 2nd April 2012, 03:19 AM   #2382
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by moschfet View Post
Hello Bob,

I bought your book and I'm thrilled. Even as a German I was able to understand the most immediately. The book has saved me a lot of work and explains many things I still wanted to know.

Thanks
Hi moschfet,

Thank you for your very kind words. I'm glad you enjoyed the book.

Cheers,
Bob
  Reply With Quote
Old 2nd April 2012, 03:08 PM   #2383
GregH2 is offline GregH2  Australia
diyAudio Member
 
GregH2's Avatar
 
Join Date: Oct 2009
Hi Bob,

Once again, many thanks for your excellent book.

I have a quick question I'm hoping you or someone else on the forums might be willing to answer. In your book you discuss several ways of implementing thermal compensation for power MOSFET amplifiers, mentioning that the standard Vbe multiplier will overcompensate. You then provide an array of slightly more complicated solutions.

Something I am wondering is whether a simple Vgs multiplier using a small mosfet could be used instead? What are your thoughts?


Your advice is appreciated!

Regards,

Greg.
  Reply With Quote
Old 2nd April 2012, 06:46 PM   #2384
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by GregH2 View Post
Hi Bob,

Once again, many thanks for your excellent book.

I have a quick question I'm hoping you or someone else on the forums might be willing to answer. In your book you discuss several ways of implementing thermal compensation for power MOSFET amplifiers, mentioning that the standard Vbe multiplier will overcompensate. You then provide an array of slightly more complicated solutions.

Something I am wondering is whether a simple Vgs multiplier using a small mosfet could be used instead? What are your thoughts?


Your advice is appreciated!

Regards,

Greg.
Hi Greg,

This is an interesting idea. I must admit that I have not tried it, or given much thought to it.

If it is used in a MOSFET amplifier where the VAS drives the MOSFETs directly (i.e., no driver), then I'm guessing it might work quite well.

If it is used in a MOSFET amplifier where BJT drivers are used, then a more complex arrangement might be needed if the Vbe TC of the BJT drivers is to be properly taken into account. Taking the BJT TC Vbe into account may also depend on whether the BJT drivers have their own heat sinks or whether they are mounted on the main heat sink along with the power MOSFETs.

Also bear in mind that the type of MOSFET has a bearing on all of this. Lateral MOSFETs are often biased at a point close to where their TC is nearly zero. Vertical MOSFETs, the main context in which I was discussing bias spreader temperature compensation, have a positive temperature coefficient of Id vs temperature that is big enough to need some temperature compensation.

Cheers,
Bob
  Reply With Quote
Old 3rd April 2012, 02:48 AM   #2385
GregH2 is offline GregH2  Australia
diyAudio Member
 
GregH2's Avatar
 
Join Date: Oct 2009
Thanks Bob.

The circuit I have in mind is a vertical fet output with folded drivers as per your figure 11.16. As I understand it, this pretty much removes the Vbe temperature variations of the drivers from the equation and leaves us with only that of the power fets. So in this situation the Vgs multiplier might be ideal?

Thanks for your advice.

Greg.
  Reply With Quote
Old 4th April 2012, 05:44 PM   #2386
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by GregH2 View Post
Thanks Bob.

The circuit I have in mind is a vertical fet output with folded drivers as per your figure 11.16. As I understand it, this pretty much removes the Vbe temperature variations of the drivers from the equation and leaves us with only that of the power fets. So in this situation the Vgs multiplier might be ideal?

Thanks for your advice.

Greg.
Hi Greg,

This is largely true, but not completely true. The folded EFs, Q1 and Q2, introduce a Vbe TC, but note that it is of opposite sign to that we normally must consider with non-folded drivers. This is a nice feature and gets us closer to where we want to be. However, Q1 and Q2 don't warm up the same (or as much) as the power MOSFETs, unless they are mounted on the main heat sink. So a Vbe multiplier bias spreader transistor may still be needed to be mounted on the heat sink. It may not need to have as much TC-inducing "strength", so it may want to be in series with a diode or two in the bias spreader that are only exposed to board ambient.

See Figure 11.8 for some useful bias spreader arrangements. See also Figure 14.11 for some useful bias spreader arrangements as well.

Cheers,
Bob
  Reply With Quote
Old 5th April 2012, 02:31 AM   #2387
GregH2 is offline GregH2  Australia
diyAudio Member
 
GregH2's Avatar
 
Join Date: Oct 2009
Thanks for your advice Bob, most appreciated. Going to go work on a suitable design now.

It's great that these forums allow us to have our questions personally answered by the authors of our textbooks!

Greg.
  Reply With Quote
Old 5th April 2012, 03:33 AM   #2388
AKSA is offline AKSA  Australia
diyAudio Member
 
Join Date: Sep 2001
Location: Melbourne, Australia
Bob,

I recently measured some large mosfets from Fairchild and found the tempco to be 6mV/C for quiescent currents up to 200mA.

As you say, and Greg and I are both working on this together in fact, the bipolar drivers would each compensate to pull back the combined 12mV/C of the large mosfets a little, but OTOH they are running at reasonably constant power and so merely reach equilibrium after twenty minutes or so and likely would not be on the output heatsink metering temperatures anyway.....

So, if we assume that around (4.1 x 2) - (0.65 x 2) = 6.9V is required at the bias generator, then another effective solution using a cheap bipolar device as bias control might be to devise a Vbe multiplier that attaches atop one of the output devices - say a BD139 with 1.8mV/C, and set it to around 11mV/C by selecting 5 times Vbe between CB - 3.15V, and giving a total C to E voltage of 3.8V. This would typically mean Rbe of say 680R, and Rcb of 3k3, npv. I would assume at least 15mA flowing through the Vbe multiplier. The difference between the required bias, 6.9V, and 3.8V of the tailored Vbe multiplier could then be taken up with a simple resistor, in this case of 206R, obviously a pot for adjustment.

Is this on the right track?

Thank you very much for your time. I have found the tempco of a large mosfet at low bias currents to be an accident waiting to happen; they can get out of control very quickly.

Cheers,

Hugh
__________________
Aspen Amplifiers P/L (Australia)
www.aksaonline.com

Last edited by AKSA; 5th April 2012 at 03:39 AM.
  Reply With Quote
Old 5th April 2012, 10:46 PM   #2389
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by AKSA View Post
Bob,

I recently measured some large mosfets from Fairchild and found the tempco to be 6mV/C for quiescent currents up to 200mA.

As you say, and Greg and I are both working on this together in fact, the bipolar drivers would each compensate to pull back the combined 12mV/C of the large mosfets a little, but OTOH they are running at reasonably constant power and so merely reach equilibrium after twenty minutes or so and likely would not be on the output heatsink metering temperatures anyway.....

So, if we assume that around (4.1 x 2) - (0.65 x 2) = 6.9V is required at the bias generator, then another effective solution using a cheap bipolar device as bias control might be to devise a Vbe multiplier that attaches atop one of the output devices - say a BD139 with 1.8mV/C, and set it to around 11mV/C by selecting 5 times Vbe between CB - 3.15V, and giving a total C to E voltage of 3.8V. This would typically mean Rbe of say 680R, and Rcb of 3k3, npv. I would assume at least 15mA flowing through the Vbe multiplier. The difference between the required bias, 6.9V, and 3.8V of the tailored Vbe multiplier could then be taken up with a simple resistor, in this case of 206R, obviously a pot for adjustment.

Is this on the right track?

Thank you very much for your time. I have found the tempco of a large mosfet at low bias currents to be an accident waiting to happen; they can get out of control very quickly.

Cheers,

Hugh
Hi Hugh,

Yes, the TCvgs of vertical MOSFETs you quote of about 6 mV/C is about what I have seen over a long time. And yes, if it is not compensated it can be a disaster waiting to happen. Those accustomed to using lateral MOSFETs must be especially mindful of this difference. The zero TC crossover point for large vertical MOSFETs is on the order of several amps.

Placing the portion of the Vbe multiplier on the case of the power MOSFET definitely helps get closer to the real heat source. The thermal resistance of the insulator in general (i.e., even for BJT amps) interferes with good thermal tracking, as does the fact that heat sinks are far from isothermal.

As I mentioned in my MOSFET power amplifier paper with EC, vertical MOSFETs are far more temperature stable when reasonably compensated than bipolars. In a sense, I guess, they lie between BJTs and laterals. I was able to show this with a plot of bias current vs time after a large change in operating signal power level.

Of course, the availability of ThermalTrak BJT transistors makes it possible to build BJT amplifiers with thermal stability just as good as those using vertical MOSFETs (maybe even better, since the sensing diode in the ThermalTraks is thermally closer to the BJT junction (whose temperature is what really counts).

You are definitely on the right track. The whole thermal circuit of temperature compensation of any output stage is so complex that some experimentation and lab measurement is still important in evaluating and establishing thermal stability.

BTW, I usually believe that we must consider both static and dynamic thermal stability. If you turn on an amplifier with no signal and plot quiescent bias as a function of time, you are evaluating static bias stability. Similarly, If you do the same with the amplifier passing a signal and putting some power into the load, you are still largely evaluating a quasi static version of bias stability.

Dynamic bias stability defines how the amplifier bias current reacts to large changes in power dissipation as a result of signal. For example, if you let the amp warm up with no signal, measure quiescent bias, then run it at 1/3 power for 10 minutes, then remove signal and measure quiescent bias as a funtion of time, you are evaluating dynamic bias stability.

The second case cited above, where the amp is turned on with signal applied and samples of bias current are taken versus time, has somewhat a combination of static and dynamic bias stability aspects to it.

Cheers,
Bob
  Reply With Quote
Old 6th April 2012, 02:13 AM   #2390
AKSA is offline AKSA  Australia
diyAudio Member
 
Join Date: Sep 2001
Location: Melbourne, Australia
Many thanks, Bob,

Wonderful information.... you highlight the static and dynamic tempco, and suggest it's best done empirically.

I'm on the case,

Thanks again!

Hugh
__________________
Aspen Amplifiers P/L (Australia)
www.aksaonline.com
  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
Another realization of Bob Cordell's THD Analyzer giulianodes Equipment & Tools 37 9th October 2013 03:48 AM
best audio amplifier book?? Bouvett Everything Else 30 13th August 2012 03:43 AM
Amplifier Design Book pixie Everything Else 27 11th June 2010 09:36 PM
Project 11.1 from Slone "High-Power Amplifier" Book Karl71 Solid State 46 6th October 2008 04:47 AM
book-audio power amplifier by Douglas Self mikee12345 Solid State 8 16th November 2003 02:16 AM


New To Site? Need Help?

All times are GMT. The time now is 12:07 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