Electronics > Power/Renewable Energy/EV's

Error amp output range of control for sync Bucks


We wish to use either the LTC7803, or the LM25148 as a synch buck. But we need also to have an output current clamp, which can clamp the current by pulling down the comp pin..(eg an opamp error amp to pull down the COMP pin via a diode)....the thing is, to do this, you first  need to know the COMP pin range of control voltages.....but neither datasheet says. Do you know  these values?



For the LTC7803, I infer from the Functional Diagram that the ITH voltage range is: normally somewhere in the range of 0.65 - 2.7V (give or take slope compensation). I also see Sleep mode comparator at 0.425V

For LM25148, I infer:
- current sense voltage (ISNS+ to VOUT) < 60mV to avoid overcurrent trip. This we could guesstimate… 50 mV nominal
- current sense gain = 10. Thus we could say a range of about 0.5V
- slope compensation present. This would increase the prospective voltage range a bit. Maybe 0.75V range? Total guess there…
- but it’s not clear what DC offset is. It’s probably not zero
- you could measure it on an evaluation board maybe? Or perhaps extract it from a SPICE model?

Unfortunately most pwm controllers which handle slope compensation internally will not answer this.

I've asked Linear Tech/ADI this same question about numerous parts in the past. They don't want customers to understand their current mode controllers in such detail, as it's proprietary IP to them. And their own LTspice models also often don't behave the same as real devices in this regard.

Doesn't look like the LM25148 datasheet is any clearer about it.

When I really need a PWM controller which is accurately modeled, I will use a simpler controller like the "unitrode" family of parts, i.e. UCxxxx or UCCxxxx). But then you will likely have to take care of slope compensation manually.

thanks, i suppose we will have to build a converter, and check the ITH voltage at no load...or light load....and progressively load it till it overloads, and see the range....then hope that all the other parts do the same thing. As you know, you need a diode in the pull down of the ITH pin...so if zero duty cycle correponds to around 1v or so then thats good..eg in ucc28c43 where the 2 internal diodes mean zero DC is around 1.4v.


[0] Message Index

There was an error while thanking
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod