hf ringing after inductor on positive halfwave. - Page 4 - 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 15th October 2012, 04:03 AM   #31
diyAudio Member
 
Sch3mat1c's Avatar
 
Join Date: Jan 2003
Location: Milwaukee, WI
Send a message via ICQ to Sch3mat1c Send a message via AIM to Sch3mat1c
I'm going to guess the ringing is, in part, the OFF-state transistor's junction capacitance and lead inductance forming a resonant stub. Only thing you can do is dampen it (not really room to install ferrite beads that'll cook anyway) or slow down the gate drive.

You could increase stray inductance and snub that, but since you already have a board, that's not really handy. Well, you could move the SMT caps or cut back the ground plane, but that's a PITA.

Tim
__________________
Seven Transistor Labs
Projects and Resources
  Reply With Quote
Old 15th October 2012, 03:40 PM   #32
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
When I place a 1nf ceramic across the lower fet shouldn't I expect a change in frequency? I didn't measure any change.
  Reply With Quote
Old 15th October 2012, 08:42 PM   #33
Pafi is offline Pafi  Hungary
diyAudio Member
 
Join Date: May 2004
Location: Budapest
Quote:
Originally Posted by switchleg View Post
I tried my best to make my probe less inductive (image 1)...
Why is it soooooo long? Why don't you use the GND ring next to the tip? Why didn't you try measuring the reference point?

BTW: probe (GND) inductance causes less problem, if you can filter the common mode current.

Square and trapesoid signals both have unlimited spectrum (the slower the transition, the less the HF content). Something resonates to it, or picks up higher freq content. This can be the probe+scope, and/or the parasitic elements of LC filter (and PCB). If you can exclude one, the rest is to be blamed.
  Reply With Quote
Old 15th October 2012, 11:01 PM   #34
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
I was trying get around soldering directly to the tip of the probe so I have to use the clip (the clip covers the gnd ring) but I think I'm going to suck it up and try it. Does anyone know a good source to buy scope probes? Its an atten ads1062cal.

Also any suggestions on a good ferrite core to try and a supplier to buy from?
  Reply With Quote
Old 16th October 2012, 03:15 AM   #35
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
Ok so i did it... directly soldered to tip and gnd ring on probe tip.... no change.
Attached Images
File Type: jpg IMG_20121015_205712.jpg (205.1 KB, 63 views)
  Reply With Quote
Old 16th October 2012, 04:02 AM   #36
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
Pafi, my scope has a limit of +-16v offset when i am in the 200mv div (10x on probe) so to view the output of the mosfets i am stuck in the 2v div... makes it almost impossible to see any ringing at all.
  Reply With Quote
Old 17th October 2012, 02:28 AM   #37
diyAudio Member
 
Sch3mat1c's Avatar
 
Join Date: Jan 2003
Location: Milwaukee, WI
Send a message via ICQ to Sch3mat1c Send a message via AIM to Sch3mat1c
Switchleg: try AC mode or using a coupling capacitor. If you're trying to zoom in on ringing on a squarewave, try the adjecent supply terminal instead.

Probe technique isn't really a problem. When you see ringing, it doesn't matter so much the exact envelope and phase, just that there's a splat of nasties there. Whether the ground lead is crossing the EMI-glowey-bits makes a bigger difference, but then, the ground plane always cuts across it somewhere, so you're going to see it everywhere at some level. It's far more important to be aware of the spacial extent of the noise, and what loops it's coming from. That will also help inform your choice of grounding point.


Probe example. A pulse generator like this:
Click the image to open in full size.

makes a pulse like this,
Click the image to open in full size.
10ns/div, actual edges probably too fast for the scope to really see.

Probing the same circuit with 6" leads on a 10x probe gives a similar waveform, with a little more 'drool' (ringing and settling after the impulse).

What may surprise you is that the same is true of a 10x probe without the ground lead attached (the circuits being grounded through the AC line alone). The waveform is a lot uglier in that case, but not like 30dB worse, as the experts may lead you to believe: more like 6dB worse, and only around frequencies harmonic to the probe length. Frequencies under 10MHz or so are quite legible.

If you need very precise measurements, you must have good shielding, good impedance matching, and common mode currents well accounted for (this includes ferrite beads on the probe cables). If you just need a qualitative "it looks better or worse", you don't even need a physical connection; a capacitive probe tip or inductive loop is plenty good, and often more useful than a clip-on probe because they provide local information that circuit nodes don't. For example, I can probe a multilayer PCB and discover nothing, yet inductively observe radiation coming from mere vias.

Tim
__________________
Seven Transistor Labs
Projects and Resources
  Reply With Quote
Old 17th October 2012, 11:15 PM   #38
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
Thank you schematic, i will check my scope though i am pretty sure its on AC. it does 160v offset with 1x and 16v with 10x. Yes i can see a bad probe setup causing poor readings but in my case I am pretty sure the ringing is happening and once its visible on the scope i can hear artifacts start coming down into the audio spectrum (THD does rise). Haven't had time to look into the past couple days but I am going to try to rule out the LC filter (already tried a smaller bodied metal film cap, now to try another inductor) and if that doesn't resolve it maybe a few smaller caps in the supply rails (10nf).

Thanks for the reply!
  Reply With Quote
Old 18th October 2012, 12:51 AM   #39
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
I can try the rail though it means I can't use my soundcard as my source since it ties gnd to the earth gnd.
  Reply With Quote
Old 18th October 2012, 02:17 AM   #40
Pafi is offline Pafi  Hungary
diyAudio Member
 
Join Date: May 2004
Location: Budapest
Quote:
Originally Posted by switchleg View Post
Pafi, my scope has a limit of +-16v offset when i am in the 200mv div (10x on probe) so to view the output of the mosfets i am stuck in the 2v div... makes it almost impossible to see any ringing at all.
What I wrote means that you don't have to search for ringing on switching node, because the ringing on output can not caused only by ringing on switching node, but by the corner! (Second derivative of the switching node.) (This is why you don't see change on ringing freq.) If you don't see ringing on sw node, then you can say there is no ringing.

But: if you see a strong ringing on output, then it may be there, or may not! I repeat: the GND connection is TOO LONG and has a too high loop area to be be sure about the origin of the ringing! AND I can see no no sign of done cross-check. You can measure a theoretically clean point (=reference point) to see if your probe is OK. Am I clear? Of course the GND of your scope must stay connected to the GND of the amp, while you connect the tip to the same point. If there is no spike can be captured (you may need an external trigger), then and only then you can say the ringing is on the output.

The probe GND connection can be made 5 mm long, 3-8 mm away from the tip. You don't have to solder, all you need is a 1.2 mm diameter wire wrapped around the gnd ring, with an axial end as long as the tip.

http://toolboom.com/nfs/content/5930...ing_nozzle.jpg
http://empegbbs.com/ubbthreads.php/u...ename/tips.jpg

You can reduce ringing by many ways, but 1 thing is sure: falling and rising edge must be similar!

Quote:
Does anyone know a good source to buy scope probes?
E-bay?

Last edited by Pafi; 18th October 2012 at 02:29 AM.
  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
Tapped HF inductor: where to get? Stabby Parts 0 20th April 2011 11:38 PM
Ringing on class d output inductor. nigelwright7557 Class D 36 25th February 2011 07:49 PM
Separate DACs for each halfwave ? Bernhard Digital Source 94 20th December 2006 06:06 AM
Cure for HF oscillation on positive cycles only RichardJones Pass Labs 4 27th June 2002 12:20 AM


New To Site? Need Help?

All times are GMT. The time now is 02:51 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