My NON-discrete SODFA class-D amp - Page 11 - 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 20th November 2005, 08:17 AM   #101
diyAudio Member
 
Join Date: Sep 2004
Location: Yahoo, USA
A couple more noise reduction tips:

A switching transistor's ungrounded heatsink can radiate e-field noise into the small signal circuitry. Ground the heatsink to the output power ground, i.e. the center point between the output stage bypass capacitors (the 470uF electrolytics). This will mend your heatsink's EMI spraying ways and convert it into an EMI sink as well (Faraday shield).

By the way, the ground for all of the small signal audio circuitry should not be connected to this center point at all. That ground should only be connected to the power circuitry at one point, the final output filter capacitor ground (C25, 470nF, in your case).

The output power ground (capacitor center point) should also be the point where you tie in the grounds from your output stage power supplies. The loop area formed between the totem pole mosfets and their bypass capacitor should be squeezed down to the absolute minimum. Place these components as close together as possible, but as or more importantly, run the traces that form this loop right next to or on top of each other so that enclosed loop area is as close to zero as possible. This means that these trace will probably not run in the shortest straight lines between points.

click here for more tips about working with driver ICs.

Regards -- analogspiceman
  Reply With Quote
Old 20th November 2005, 11:29 AM   #102
Pierre is offline Pierre  France
Banned
 
Join Date: Nov 2004
Location: Paris
Chris:
I agree in that older mosfets can be more rudged because thay are a bit slower. However, IRF640N, for example, is not what I call slow: it has 67nC max gate charge and 1.16nF input capacitance, not very high IMHO. My circuit uses 6.8ohm gate resistors, and the switching waveform has 45ns rise/fall times.
NTP35N15, that I also tried, has double gate charge and capacitance, and they were quite a bit more fragile however ?

Analogspiceman:
I use a kind of alluminium plate that serves both as the heatsink for the mosfets and as the "support" for the amp board. I also realized that this heatsink, which was floating, was radiating a lot of EMI due to capacitive coupling via the isolator of one switching mosfet. So I decided to connect it to GND with a 100nF SMD capacitor, at the power side of the board. Do you think this is correct or I can do it better in order to reduce EMI emissions?

Best regards,
Pierre
  Reply With Quote
Old 20th November 2005, 12:26 PM   #103
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Hi Pierre,

The IRF640 I used, dont' think it was an N version.

There's still other factors to contend with as you mentioned before.

IRF510's took the same beating but didn't last 1/10 the time the 640's did. At the time it still lasted alot longer than an FDP3682 would.

Such Fet's take very little to drive and tend to overheat alot more readily, under less than ideal conditions.

In a P2P type setup we aren't talkign SMD devices so, what I did drill a hole in my heatsink close to the mosfet and bolted a small value ceramic cap to ground.

You can actually feel the difference it makes with your finger!

Regards,
Chris
  Reply With Quote
Old 20th November 2005, 06:45 PM   #104
Pierre is offline Pierre  France
Banned
 
Join Date: Nov 2004
Location: Paris
Quote:
In a P2P type setup we aren't talkign SMD devices so, what I did drill a hole in my heatsink close to the mosfet and bolted a small value ceramic cap to ground.
Sorry, I don't get you. Are you talking about my question of what to do to reduce EMI by coupling the heatsink to GND by means of a cap?
What difference do you feel with your finger exactly?

Best regards,
Pierre
  Reply With Quote
Old 20th November 2005, 06:58 PM   #105
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Quote:
Originally posted by Pierre


Sorry, I don't get you. Are you talking about my question of what to do to reduce EMI by coupling the heatsink to GND by means of a cap?
What difference do you feel with your finger exactly?

Best regards,
Pierre

No no just saying how it might relate to Sanders P2P implementation. As to yours, UCD uses a cap as well. It's only high frequency we're concerned with I'd imagine a cap would be fine, otherwise you have the inductance of the wire... using both could create a tank..

The difference? Without the cap my finger tingles, with the cap it doesn't!!

Regards,
Chris
  Reply With Quote
Old 20th November 2005, 08:24 PM   #106
Pierre is offline Pierre  France
Banned
 
Join Date: Nov 2004
Location: Paris
Sorry, I still don't understand what capacitor are you talking about...
Thanks for the clarification about heatsink decoupling one.
  Reply With Quote
Old 20th November 2005, 08:40 PM   #107
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Quote:
Originally posted by Pierre
Sorry, I still don't understand what capacitor are you talking about...
Thanks for the clarification about heatsink decoupling one.
It's ok .. considering how I typed it.

Yeah I was talking about bypassing the heatsink to ground. You have a PCB so you used a surface mount cap, but if you wire p2p you can't really use surface mount. So for p2p I drilled a hole in the heatsink near the mosfet and screwed a through hole cap to it, as close to the mosfet as possible.
  Reply With Quote
Old 20th November 2005, 09:47 PM   #108
Pierre is offline Pierre  France
Banned
 
Join Date: Nov 2004
Location: Paris
Oh, yes, thanks, I had understood that part
I was wondering if the cap that you say that makes so big difference (corroborated by your finger) is this one or you were talking about some other cap.
  Reply With Quote
Old 20th November 2005, 09:55 PM   #109
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Quote:
Originally posted by Pierre
Oh, yes, thanks, I had understood that part
I was wondering if the cap that you say that makes so big difference (corroborated by your finger) is this one or you were talking about some other cap.

yep.
  Reply With Quote
Old 21st November 2005, 09:36 AM   #110
SSassen is offline SSassen  Netherlands
diyAudio Member
 
SSassen's Avatar
 
Join Date: Mar 2005
Location: Edam, The Netherlands
Subw01, analogspiceman, Chris, Pierre,

Thanks for all the tips and suggestions, let me summarize them below:

Quote:
You may also want to set Vcc on the IR2011 to 15v so that during peak positive output of the amp when the low time may be be very short, the upper capacitor, C20, will maintain a charge of at least 11.3v when it draws charge off of the lower one, C19. When it does so, it temporarily could take C19 from 15 down to 12 volts. But, that worse case scenario of the bootstrap capacitor dropping the lower one to 12v could happen only if the bootstrap one is zero when it draws charge from the lower one.
Hence, I'm better off running the IR2011 with a 15-volt supply.

Quote:
Step 1. Get yourself a ferite core through which you can pass three to five turns of your 'scope probe's cable (the coax). Some of the larger, split-and-polished, clip-on anti-EMI test beads work nicely for this; so do ungapped ring cores with center holes large enough to pass the 'scope probe connector. Try to get a core with a cross section of at least half a square centimeter. Wrap the turns so that the core ends up next to the oscilloscope rather than the probe. Now rerun the ground noise pollution test from before. Common mode noise should be reduced by an order of magnitude or more.
Good thinking, I got clip-on EMI-reduction beads that fit the bill exactly.

Quote:
A switching transistor's ungrounded heatsink can radiate e-field noise into the small signal circuitry. Ground the heatsink to the output power ground, i.e. the center point between the output stage bypass capacitors (the 470uF electrolytics). This will mend your heatsink's EMI spraying ways and convert it into an EMI sink as well (Faraday shield).
Excellent, it indeed is not connected to anything currently, I'll drill a hole into it and connect it to GND.

Quote:
By the way, the ground for all of the small signal audio circuitry should not be connected to this center point at all. That ground should only be connected to the power circuitry at one point, the final output filter capacitor ground (C25, 470nF, in your case).

The output power ground (capacitor center point) should also be the point where you tie in the grounds from your output stage power supplies. The loop area formed between the totem pole mosfets and their bypass capacitor should be squeezed down to the absolute minimum. Place these components as close together as possible, but as or more importantly, run the traces that form this loop right next to or on top of each other so that enclosed loop area is as close to zero as possible. This means that these trace will probably not run in the shortest straight lines between points.
Alright, looks like I'll be doing a new perf. board layout afterall.

I've also thought myself how to use LTspice more proficiantly over the weekend, so I'll be modelling this design with more than just my calculator this week to see whether there's parts of the design that could use some further refining. I also stumbled across the following, compare the below image to the one I posted recently.

LT1016 ringing due to inproper termination
Click the image to open in full size.

LT1016 output in my design
Click the image to open in full size.

This is exactly what I'm seeing at my comparator outputs, hence we could be looking at reflections due to the outputs not being properly terminated. I'll add 470-ohm resistors to ground to see whether that helps.

Best regards,

Sander Sassen
http://www.hardwareanalysis.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
Noise problems in non discrete class-d amp Baldin Class D 26 18th February 2007 09:00 AM
Discrete Class-D anno 1986 Baldin Class D 11 26th October 2006 08:30 PM
My all discrete prefilter class-D kartino Class D 10 7th August 2006 09:47 AM
My 13x discrete class-D kartino Class D 6 21st November 2005 08:44 AM
recommend discrete Class D rellik Class D 11 3rd November 2005 02:27 AM


New To Site? Need Help?

All times are GMT. The time now is 02:33 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