poor switching - Page 2 - diyAudio
Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

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 June 2004, 10:00 PM   #11
Pafi is offline Pafi  Hungary
diyAudio Member
 
Join Date: May 2004
Location: Budapest
The problem is saturation (at 50 ohm drive), and insufficient basis current (at 1k).

Look at this: http://www.hszk.bme.hu/~sp215/elektro/stuff.gif

Or this one: http://www.hszk.bme.hu/~sp215/elektro/stuff2.gif

The 10 nF is simulating the gate capacitance.

But why don't You use driver IC?
  Reply With Quote
Old 12th June 2004, 10:01 PM   #12
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Quote:
Originally posted by janneman
Well, my graph showes 8V across 11 ohms from a 9V battery, which for a small signal transistor is good (VCEsat=1V), and the switching speeds show that it works. It's really a trivial cct of course. If it doesn't work in your case, it's not the transistor, believe me.

Did you try with the puls gen against ground? How did you check that the floating puls gen works?


Jan Didden
Yeah I said it looks good, but it doesn't get the job of sourcing an amp done, is all that I meant. This isn't government work According to my simulation, you only get 375mA out of it.

The floating pulse gen. works..it's the green line in the graphs I posted.

Thanks,

Chris.
  Reply With Quote
Old 12th June 2004, 10:17 PM   #13
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Quote:
Originally posted by Jaka Racman
Hi Chris,

I do not know if your requirements are really serious (10Mhz switching), but presuming that they are I do not think you have chosen the right way. I have been simulating switching designs almost to death, and I have found out that component models usually do not work very well in Spice. Sometimes FET models lack package parasitic inductances etc. So for longer simulations (over many switching cycles) I prefer to use S (switch element).

If your requirements are really 10MHz, you are candidate for IXYS RF (former Directed Energy) FETs. Also for 10Mhz switching, you would not use classic drive circuit, but some kind of resonant driver. Those usually use gate inductor instead of gate resistor. Such drivers have low losses, since they conserve energy in resonant circuit instead of dissipating it in the gate resistor. There are many variants of such designs, and I may try and find some examples, if you are interested.

Best regards,

Jaka Racman
Hi Jaka,

Yeah I've studied IXY's app notes thoroughly back when I was dreaming of switching DSD to the load directly at high power..and came to the conclusion that kind of a circuit is a bit out of my grasp at this point, but its' good reading!

I make no secret of the fact I'm working on a high power UCD circuit....500 watts continuous is my goal for the half bridge. I won't build what I can't simulate..

I've had the exact circuit simulate a THD of .2% with no extra filtering using the 1N5401's to drive these very mosfets..but I did calculate 1 amp at 10Mhz in order to ensure very snappy switching at the modulation extremes...1%.....99%...how fast would be required? The idea behind that being a good margin of safety for the math, if it could switch on at 100ns, it would work great at 350khz or so...and doubling up the power mosfets would have been trivial with 1 amp drive.

Thought the zetex would be better. In the circuit itself it's just ugly, and the simple switch test I did shows it rather slow for some reason.

The S-switches...I'll have to figure out how to use those one day..would make the proof of concept playing around alot easier.

There's a few other glaring problems that still need working out, but I thought I'd try and nail down the output aspect of it first.
Thanks

Chris.
  Reply With Quote
Old 12th June 2004, 10:24 PM   #14
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Quote:
Originally posted by Pafi
The problem is saturation (at 50 ohm drive), and insufficient basis current (at 1k).

Look at this: http://www.hszk.bme.hu/~sp215/elektro/stuff.gif

Or this one: http://www.hszk.bme.hu/~sp215/elektro/stuff2.gif

The 10 nF is simulating the gate capacitance.

But why don't You use driver IC?
Hi,

Two good reasons, my lab consists of my computer..they're impossible to simulate, and rarely find models of them, the driver you mentioned isn't one I've seen before, but it sure looks promissing.

The main reason is...it goes against the UCD design..of symetrical non saturated very fast transistors. I don't think a driver chip would be an improvement on that...even if I could get a working simulation going of it.

Once I get the simulations as perfect as I can, I plan on just prototyping it up and watching it smoke

I don't even have a scope

Thanks alot for your input, it's most welcome and right on the money.

I don't think my calculations are off in my little switch example..and I did use half the beta to calculate with than its' rated for at that current...So I thought it would work...but since it doesn't...it seems to fully switch with a base resistance of around 330....and will have a baker clamp..

Maybe I'll play with it more later

Cheers
Chris
  Reply With Quote
Old 12th June 2004, 11:05 PM   #15
Pafi is offline Pafi  Hungary
diyAudio Member
 
Join Date: May 2004
Location: Budapest
Chris!

Please tell me what does "UCD design" mean! I'm not very good at terminology in English!

"I don't think a driver chip would be an improvement on that"

These are technically perfect, I think:

Features
• BiCMOS/DMOS Construction
• Latch-Up Proof: Fully Isolated Process is Inherently
Immune to Any Latch-up.
• Input Will Withstand Negative Swing of Up to 5V
• Matched Rise and Fall Times ............................... 25ns
• High Peak Output Current .............................. 9A Peak
• Wide Operating Range .............................. 4.5V to 18V
• High Capacitive Load Drive ........................... 47,000pF
• Low Delay Time ........................................... 30ns Typ.
• Logic High Input for Any Voltage from 2.4V to VS
• Low Equivalent Input Capacitance (typ) .................7pF
• Low Supply Current .............. 450ľA With Logic 1 Input
• Low Output Impedance ........................................ 1.5Ω
• Output Voltage Swing to Within 25mV of GND or VS

It's in SO-8 package. Do we need more? You can replace it with ideal comparator+1,5ohm resistor in simulation!

But if you insist the simulation, then take this: http://www.hszk.bme.hu/~sp215/elektro/stuff3.gif

I've used this untill I met MIC4421. (But with FDT457-458)

Note: driver dissipation should not exceed one of the output device!

P.S.: Scope, and Laplace-transformation are the best friend of engineer!
  Reply With Quote
Old 12th June 2004, 11:31 PM   #16
diyAudio Member
 
Join Date: Dec 2002
Location: Slovenia
Hi Chris,

now that I know your intentions you have my best wishes. I have been doing SMPS design professionally for many years, but I would not dare to design anything switching without a scope, unless using a previously proven design or blocks. It might be a professional deformation, but I feel naked without a scope. On the other hand I know a very bright guy (he is now a university professor, but he could teach some professors when he was still a student) that can do gigahertz ham radio designs only with theory and analog multimeter. Maybe you could post your schematics for comment before you actually start building?

Best regards,

Jaka Racman
  Reply With Quote
Old 13th June 2004, 12:34 AM   #17
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Hi,

Pafi, thanks for all your help.
UCD stands for "Universal Class D" formely "Ultimate Class D" and is the best class D topology out there today (mueta doesn't count yet, and probably won't sound as good either)..

I will find you the link to the patent info so you too can see it's elegant beauty.

Bruno's UCD patent

Have a look at that and you'll understand why I don't want to use a driver chip. Nothing is stopping me from doing it, but I dont' think it would be as good as what's on the patent.

Believe me guys, I'd love to have a scope, and for this type of project, naked without it is spot on, probably even more so like trying to land a jet on an aircraft carrier in the middle of a winter storm with half the landing gear down. Why should that stop me from trying?

I'll P2P a prototype, power it with +-12 regulated 1 amp...see what smokes....tweak it a little..if it works..I'll work on the power supply...Maybe I can take a local repair shop hostage for a few hours and use theirs? See what happens.

In the meantime I can perfect the design as much as I can on spice, the better it gets the more confidence I'll have with making a prototype. If the prototype works half decent..(it will either smoke, play, or something in between) I'll make the pcb..hopefully by then I'll have a scope at least anyway.

It's a learning experience, keeps me busy, and if I should wind up with a great amp when it's all said and done..I'll be happy.

As far as posting the schematic goes, once I've taken it as far as I can I'd be happy to. I'd only be laughed at if I posted what I had now..

Thanks for the kind words and help, I won't be giving up anytime soon.

Do you guys think I made the right choice of transistor after all?
I'll be looking up the ones Pafi mentioned later once I've had some rest.

I reaaaally should have a scope to experiment with what makes everything work best for this sort of thing..That will have to come later, but there's no reason why I can't have a good design of it now. I have a dead monitor I keep looking at sideways hmmmmm.

Take care, and thanks again.

Chris
  Reply With Quote
Old 13th June 2004, 01:03 AM   #18
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Hi,

This is my incentive to continue, and to show that it can be done, you can see it has a problem or two..none the less....enjoy.

Chris
Attached Images
File Type: gif needswork.gif (92.9 KB, 162 views)
  Reply With Quote
Old 13th June 2004, 11:54 AM   #19
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
I draw this some time ago thinking in fast MOSFET drive for low voltage SMPS, but I think there are also some useful ideas for class D

Click the image to open in full size.


This shows the latter circuit in action, driving an IRFZ48V from simple BC337/327 bipolars and 10ohm Rg in a push-pull circuit [turn-off from conducting 60A or so]. Leftmost trace is Vgs at 5V/div [I use 18V drive], rightmost trace is Vds at 10V/div, time is 200ns/div and gate is discharged in less than 120ns. This also shows that high current MOSFETs are not perfect and suffer some degree of 'current tail' and 'storage time' effects as opposed to what common wisdom says

Click the image to open in full size.

In the other hand, with ZTX851/951 bipolars configured as a simple class-B buffer driven from paralell CMOS buffers I've got performances as good as 50ns or less charge/discharge times for an IRFP460 switching 300V [including 'miller' capacitance effects, very noticeable at these high voltages]
  Reply With Quote
Old 13th June 2004, 12:32 PM   #20
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
I forgot to say that in the latest version of that circuit I reduced Rg to 4.7 ohms and got 70ns discharge time without exceeding the 1A peak BC327/337 current limit [voltage drop on Rg showed about 650mA peak]
  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
Poor Man's Strad peace brainerd Multi-Way 75 14th August 2012 08:53 PM
poor man reciever v1d9uy Solid State 4 6th January 2006 03:28 PM


New To Site? Need Help?

All times are GMT. The time now is 12:15 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright Š1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2