Sg3525a ic.

Thread Starter

shortbus

Joined Sep 30, 2009
10,045

iONic

Joined Nov 16, 2007
1,662
Yes, it's a bit confusing to me as well. It does state the freq rang from 100Hz to 400kHz and there is mention of duty cycle on the bottom of page 4... DCmin 0%, DCmax 49%.
A circuit the shows some PWM adj. is in figure 2. Lab Test Figure.
But other than that I am as confused as you are....Didn't read the entire DataSheet yet. I was looking for an application note for the chip but was unable to find one.
 

praondevou

Joined Jul 9, 2011
2,942
The pulse width depends on the chosen frequency (according to datasheet ranges from 100Hz to 400kHz) and the duty cycle.
Deadtime is the time the chip makes sure that both output pulses are off. This is to avoid cross conduction in half/full bridge configurations.

If you want to use the 3525, I found this project which seems to be exactly what you are looking for.
 

Thread Starter

shortbus

Joined Sep 30, 2009
10,045
@iONic- Thought you might chime in :)

@praondevou- I saw that circuit linked, but the link didn't work for me, yours does, thanks. That was always my understanding of dead time too. If you can't adjust the pulse width separate from frequency, how come they call the chip a PWM? Thought the ability to adjust pulse width was the purpose of PWM?
 

SgtWookie

Joined Jul 17, 2007
22,230
There are quite a few things you need to understand about switching supplies in order to be able to use the IC.
Here's just a starter; voltage vs current mode feedback: http://focus.ti.com/lit/an/slua119/slua119.pdf

Here is a list of selected applications notes:
http://focus.ti.com.cn/cn/lit/ml/slub007/slub007.pdf

Here is ST Microelectronics' application note for the SG1525/SG3525 SG1527/SG3527 family:
http://www.st.com/stonline/books/pdf/docs/1689.pdf

TI's datasheet for the UCx525, UCx527 family:
http://focus.ti.com/lit/ds/symlink/uc1525a.pdf

These IC's have actually been around for a good while. I just mentioned them because the outputs could be made TTL compatible, and you could probably salvage one from a computer power supply.

I've seen a motor controller using a 3525 somewhere but can't remember offhand.
 
Last edited:

praondevou

Joined Jul 9, 2011
2,942
If you can't adjust the pulse width separate from frequency, how come they call the chip a PWM?
Of course you can adjust the pulse width separate from frequency, but by now you probably know this if you had a look at Sgt.Wookies links.:)

There are variable frequency systems too, but in general the designer determines a frequency he wants to work with, this one will be fixed and the pulse width depends on the duty cycle (+dead time) which in turn depends on the configuration of the IC, it's regulation (which depends on its feedback) and so on.

The frequency you choose depends on a number of parameters including max allowable dissipation of the switching device, generated noise level, power transformer characteristics.

With motors it is also a common practice, in case the switching devices like mosfets etc are getting to hot, to decrease the switching frequency in steps, thus lowering switching losses.
 
Last edited:

Thread Starter

shortbus

Joined Sep 30, 2009
10,045
Thanks to everyone for the answers, got some more reading to do.

My thoughts were to use one of these to switch a low side mosfet gate for a easy PWM motor drive.
 
Top