what do you think about these gate drivers??

ttako · 06:10 PM

Hi,

What do you think about the Silabs Silabs Si841x series isolators (<10nsec propagation delay) plus the MAX 15024 Driver (<20nsec propagation delay, 4A peak)?
Or the Silab 8233/34 integrated solution (it has some more delay, like 50 nsec)?

Tamas Tako

Pafi · 10th October 2009, 07:06 PM

Digital isolators have a huge jitter, they make significant noise. 5 ns-> -66 dB noise in open loop, approximately -80...-90 dB in closed loop.

sawreyrw · 10th October 2009, 08:24 PM

pafi,

Please explain how 5ns of jitter causes -66db of noise. What PWM frequency and closed loop bandwidth are you assuming in you calculations?

Rick

Pafi · 10th October 2009, 09:29 PM

Rick!

2.5us=0dB, 5ns=-54 dB, but the hearable part is ~12 dB lower (400kHz/20kHz, linear power distribution). Maybe 6 dB lower or higher, it doesn't matter, it's too much all the way.

I didn't care closed loop bandwidth. The gain of the closed loop (at ~10 kHz) is what counts. (It's a simplification, I know, but it doesn't worth the effort of the exact calculations.)

ttako · 11th October 2009, 09:28 AM

Dear Pafi,

If I remember correctly, the jitter of this particular type is below 1nsec, in fact I think I've read somewhere its 250 picosec......

So, do you still think it is a bad solution?

Thanks,

Tamas

Pafi · 09:57 AM

Tamás!

I think this specification can be misleading, they measured it with a single constant freq, but I'm quite sure that at different freqs (or pulse widths) the delay is different. (RF modulation freq is only 700 MHz, that's the source of my doubts.) I think this is the relevant specification:

"Precise timing (typical)
1.5 ns pulse width distortion"

But try it in a simple circuit! Maybe I'm wrong.

10th October 2009, 06:01 PM	#1
ttako diyAudio Member Join Date: Dec 2002 Location: Hungary/ Budapest	what do you think about these gate drivers?? Hi, What do you think about the Silabs Silabs Si841x series isolators (<10nsec propagation delay) plus the MAX 15024 Driver (<20nsec propagation delay, 4A peak)? Or the Silab 8233/34 integrated solution (it has some more delay, like 50 nsec)? Tamas Tako Last edited by ttako; 10th October 2009 at 06:10 PM.

11th October 2009, 09:54 AM	#6
Pafi diyAudio Member Join Date: May 2004 Location: Budapest	Tamás! I think this specification can be misleading, they measured it with a single constant freq, but I'm quite sure that at different freqs (or pulse widths) the delay is different. (RF modulation freq is only 700 MHz, that's the source of my doubts.) I think this is the relevant specification: "Precise timing (typical) 1.5 ns pulse width distortion" But try it in a simple circuit! Maybe I'm wrong. Last edited by Pafi; 11th October 2009 at 09:57 AM.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
help with gate drivers(IR2110)	maverick_123	Class D	7	4th December 2007 08:53 PM
Models for IRF gate drivers	tlf9999	Class D	1	23rd October 2005 08:06 AM
Gate drivers	Jabbaz	Class D	12	8th January 2005 02:03 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	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

10th October 2009, 07:06 PM	#2
Pafi diyAudio Member Join Date: May 2004 Location: Budapest	Digital isolators have a huge jitter, they make significant noise. 5 ns-> -66 dB noise in open loop, approximately -80...-90 dB in closed loop.

10th October 2009, 08:24 PM	#3
sawreyrw diyAudio Member Join Date: Apr 2006 Location: Minnesota	pafi, Please explain how 5ns of jitter causes -66db of noise. What PWM frequency and closed loop bandwidth are you assuming in you calculations? Rick

10th October 2009, 09:29 PM	#4
Pafi diyAudio Member Join Date: May 2004 Location: Budapest	Rick! 2.5us=0dB, 5ns=-54 dB, but the hearable part is ~12 dB lower (400kHz/20kHz, linear power distribution). Maybe 6 dB lower or higher, it doesn't matter, it's too much all the way. I didn't care closed loop bandwidth. The gain of the closed loop (at ~10 kHz) is what counts. (It's a simplification, I know, but it doesn't worth the effort of the exact calculations.)

11th October 2009, 09:28 AM	#5
ttako diyAudio Member Join Date: Dec 2002 Location: Hungary/ Budapest	Dear Pafi, If I remember correctly, the jitter of this particular type is below 1nsec, in fact I think I've read somewhere its 250 picosec...... So, do you still think it is a bad solution? Thanks, Tamas