BenBoost questions

[tsteele93]

I have some questions about the max current on the BenBoost. I am curious how 1.2A was arrived at as a max? Was it just through testing, or by design.

More specifically, I am curious if the original idea was that the driver might run higher currents but it just didn't pan out in actual real world implementation.

If it was planned that the driver handle higher currents, but it didn't work out to do so, then has any testing been done to try and figure out what the weak link(s) is/are that restrict the max current?

I was reading some driver design articles and came across something on linear's site that made me wonder if the Schottky's 2A max avg current could be a limitation. The article suggested a 3A max average for a higher power LED driver. That was for a 900mA driver.

My ultimate question is, do you think using a higher current Schottky would help the driver to handle more current?

The reason I ask, is that I think that I found a Schottky that looks like it would handle 50% more current and twice the voltage for not too much more ($0.31 vs $0.47) if it would make an actual difference in the capability of the driver.

It also has a Vf of 460mV @ 3A which is the same as the original 2A/10V version.

DIODE SCHOTTKY 20V 3A 123-SOD - RB051M-2YTR - $0.47

Any thoughts?

:thanks:

 

[AnthoT]

I've had success using the LM3410 with no schottky and I hit 1300mA... After that thermal protection kicked in real fast... I've got videos of it on my site if you'd like proof :beer:
But using a higher rated shottky can't make it worse only better I beleive

 

[tsteele93]

I've had success using the LM3410 with no schottky and I hit 1300mA... After that thermal protection kicked in real fast... I've got videos of it on my site if you'd like proof :beer:

But isn't the idea to NOT hit thermal protection?

But using a higher rated shottky can't make it worse only better I beleive

Shouldn't hurt anything but the wallet, but do you think it will help or am I barking up the wrong tree? Or is there somewhere else in the circuit that is the weak link?

 

[AnthoT]

Generally speaking you'll never hit the switch current, most of the time not even 50% of it... So 1.2-1.3 Amps is really the max for the LM3410. so you would be wasting your money on the schottky...

 

[cilegray]

The limiting factor in your driver, in this case, is your driver.

IIRC I believe it was benmwv that did the testing but as per datasheet, it would have a maximum rated output of 1.26A @ 5V, and that's just a quick calculation assuming 100% efficiency (2.1A @ 3V)

 

[tsteele93]

The limiting factor in your driver, in this case, is your driver.

IIRC I believe it was benmwv that did the testing but as per datasheet, it would have a maximum rated output of 1.26A @ 5V, and that's just a quick calculation assuming 100% efficiency (2.1A @ 3V)

Ok, that is what I was looking for... Thanks for saving me money!

I guess that I had the feeling that this driver was supposed to do more current but wasn't hitting it because of over temp.

So basically heat sink it like mad and keep it under an amp and parallel them if you need more...

:thanks:

 

[benmwv]

AnthoT, I'm calling BS on you running it without a schottky. That diode is as key to a boost driver as the inductor is.

Anyways Tsteele, the limiting factors are switch current and input voltage. Im sure the driver could do 1.5, even 1.6A with a fresh battery and the switch current maxed out, but when the voltage drops just a little and the switch current is still maxed the overall input power drops and it can no longer achieve the same currents at a certain voltage.

Also more current = more heat and as it heats the efficiency goes down (means more heat), which makes the output go down.

 

[AnthoT]

AnthoT, I'm calling BS on you running it without a schottky. That diode is as key to a boost driver as the inductor

Really? I'm not joking I've made a few using no shottky. I'll add one to my next design then.... I never noticed any issues without it so far. Thanks for letting me know :beer:

 

[benmwv]

What did you do just short the pad? I don't see how that could possibly work.

 

[AnthoT]

What did you do just short the pad? I don't see how that could possibly work.

I didn't even have a pad for it.. I used a bit of a different design here you could see AtMa-BoostDriver

I got some videos of that exact one on the page as well... It did work. But I changed my design anyway to include it :wave: so thanks for the heads up :beer:

 

[benmwv]

Put it on a scope and check the output without the schottky. It will be crazy.

I don't know if you have tried that on a real diode but you shouldn't.

Hooked to a test load the schottky is basically replaced by the test load diodes. This means the test load diodes block the reverse spikes from the inductor instead of the schottky. if you hook it up to a laser diode those spikes will fall on the diode (which can't stand much reverse voltage like a regular diode can). Also you need to have caps after the blocking diode to smooth the output, you can't do that if you are using the LD as the inductors diode.

 

[AnthoT]

I just killed a diode trying this but here you go
My first LED'd diode | Flickr - Photo Sharing! that diode is LED'd now

I did this yesterday afternoon. It lit the diode but the diode is LED. I also should not have powered the diode without a module :banghead: oh well this is a lesson learned for me :beer: