Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

SuperBoost Drive (5.5+ amp boost driver)






Hiemal

0
Joined
Dec 27, 2011
Messages
1,443
Points
63
Re: SuperBoost Drive

So, some more testing was done, and it seems I was a bit too optimistic about the capability of this driver design, or my batteries are absolute sh*t.

The maximum amount of current I can seemingly pull from the driver with the current batteries I have, seems to be about 2.7-2.8 amps. This is boosting to 5.2 volts, roughly the voltage required for a 9 mm 3 watt 445 nm diode.

The current draw from the batteries is about 5-6 amps. Their voltage sags a TON, from 4.1 volts to 3.5 volts... So that definitely leads me to believe I simply cannot test the maximum current capability of this driver without getting something better to power it with. The driver just sucks the batteries dry before I get a chance to see what it's capable of... 18650s don't cut it here I guess, or mine can't anyway!

I also made some slight modifications to the design of the driver (had to swap the inductor, remove a cap) and it works considerably better now, just produces a decent amount of heat at the currents flowing through it.

I'm just kind of at a loss at the moment. I don't own a high current adjustable PSU so... I don't know what to do?
 
Last edited:
Joined
Jan 29, 2014
Messages
12,031
Points
113
Re: SuperBoost Drive

Are you measuring right at the battery when you see the voltage drop that far down, or at the device? Check the voltage drop between the battery and the device, if you haven't already done so, a simple thing which might be an insult to suggest it :p but thought it worth a mention. If need be, I may be able to send you a DC power supply which will work for you in trade :) which may run the device fine, but the real test would be with a battery we would use with it, otherwise, what is the point.
 
Last edited:
Joined
Jul 16, 2010
Messages
107
Points
18
Re: SuperBoost Drive

Hey, LeQuack I could send you some 26700's if you need them. I mistakenly bought them a while ago and I still haven't found a use for them, so they might as well go to a good cause.
 

rhd

0
Joined
Dec 7, 2010
Messages
8,475
Points
0
Re: SuperBoost Drive

I know I'm a broken record on this point, but it never, ever, makes sense to tease a switching driver at the design stage. It's tempting, because you make a great design in Eagle and naturally want to show it off. But with switching drivers, when it's a new IC and design you haven't tested before, you need a couple revisions to work out the kinks.

People also consistently fall into the trap of basing their expectations in the stated max switch current of the IC. In reality, you have to build in a ton of efficiency loss, voltage drops, cell sag (which is normal) and usually, an additional buffer for cosmic rays and random voodoo.

On a related point, what are you dropping voltage across on your test load that doesn't vary voltage as current changes?
 

Hiemal

0
Joined
Dec 27, 2011
Messages
1,443
Points
63
Re: SuperBoost Drive

Are you measuring right at the battery when you see the voltage drop that far down, or at the device? Check the voltage drop between the battery and the device, if you haven't already done so, a simple thing which might be an insult to suggest it :p but thought it worth a mention. If need be, I may be able to send you a DC power supply which will work for you in trade :) which may run the device fine, but the real test would be with a battery we would use with it, otherwise, what is the point.

Yeah, I measured the voltage drops across my wires. Nothing overly significant, but I may try adding some additional beef to them to reduce that variable. Can you let me know what kind of current the DC power supply can handle?

Hey, LeQuack I could send you some 26700's if you need them. I mistakenly bought them a while ago and I still haven't found a use for them, so they might as well go to a good cause.

I might have to take up your offer on that. I'll do some more testing sometime today with fresh 18650's. See how they perform.

I know I'm a broken record on this point, but it never, ever, makes sense to tease a switching driver at the design stage. It's tempting, because you make a great design in Eagle and naturally want to show it off. But with switching drivers, when it's a new IC and design you haven't tested before, you need a couple revisions to work out the kinks.

People also consistently fall into the trap of basing their expectations in the stated max switch current of the IC. In reality, you have to build in a ton of efficiency loss, voltage drops, cell sag (which is normal) and usually, an additional buffer for cosmic rays and random voodoo.

On a related point, what are you dropping voltage across on your test load that doesn't vary voltage as current changes?

Oh, I know, and I thought I had made that relatively clear on the first post that the numbers I were teasing were just possibilities.

And there's no "max switch current" for this IC, since it's just a controller and a mosfet, kind of like the IC you've been using for your buck drivers. And yeah, I'm getting the efficiency losses voltage drops and cell sag all that is part of the equation, especially when dealing with such high currents.

I actually had to switch the inductor because the original one was getting so hot it was desoldering itself from its pads because the ESR was too much.

Honestly, right now I just think it's purely a limit of my power supply. Lazeerer's SXB drivers state the same, you need good cells otherwise the driver won't be able to give you the juice you're asking of it.

I mean even just looking at my 18650's datasheets, the maximum current draw for them is about 2C, which is in line with the current I was drawing from them. So.

Oh, and for these purposes I was just using a standard test load; if you're asking about the adjustable test load I was talking about, the first version of it is an op amp, voltage reference and a transistor; the op amp makes sure the transistor has a specific voltage drop based off of the voltage ref.

So, the current can change but the v-drop will always remain the same. I do have some other ideas of how to improve it (hopefully no 9 volt cell will be needed) because one is right now)...
 

rhd

0
Joined
Dec 7, 2010
Messages
8,475
Points
0
Re: SuperBoost Drive

For testing drivers, the best tool you can invest in is a $20 Lipo. Bench PSUs behave strangely with some switching drivers and cylindrical cells have their limits. But grab a 2,000 mAh 20C lipo, and you'll have no issue powering your drivers, and it will recharge quickly.
 

Hiemal

0
Joined
Dec 27, 2011
Messages
1,443
Points
63
Re: SuperBoost Drive

So some more testing was done today... I tried my best to remove as many variables as I could and managed to get the maximum current output at 5 volts to 3.3 amps.

This was still murdering my batteries. 4.2 volts dropping to 3.7v at the batteries themselves,, and 3.6 at the actual driver. Seems even the smallest variation in voltage that gets to the driver causes a pretty big change in the amount of current it can handle. Oh, and heatsinking. Very very important. Output drops a good amount unless you heatsink it nicely. Thankfully due to the design you only need to heatsink one side. The inductor and the diode.


Beginning to think the 5 amp estimate I made before is way too optimistic. :yabbem:

And the adjustable test load I made? Doesn't work with boost drivers, it seems. The op amp becomes confused and causes the v-drop to oscillate a bit. I've tried adding caps and whatnot to various places but it doesn't seem to change much except the FREQUENCY of the oscillations. Maybe need some bulk capacitance, like more than 22 uF. We'll see!...
 

Hiemal

0
Joined
Dec 27, 2011
Messages
1,443
Points
63
Re: SuperBoost Drive

I hate to double post, but I've decided to do a pretty large revision to the boards. I've made room for a much heftier inductor (lower ESR and better current handling ability) and moved the MOSFET to allow heat to transfer a lot easier, and widened a lot of the traces on the boards. Already ordered the parts and the boards so it'll be about two weeks before I report back with the new ones.

I did, unfortunately, have to nix the reverse polarity protection mosfet. At the currents required it was another node of loss that was not needed.
 
Last edited:
Joined
Jan 14, 2011
Messages
3,816
Points
63
Re: SuperBoost Drive

About your fully adjustable test load - with just a MOSFET, a resistor, an op-amp and some other parts, it won't work. I've tried it before, and unfortunately, there's something about the high-frequency nature of the IC that throws off the op-amp or something. I remember a while ago I posted about this, but it was probably over a year ago at this point.

One thing that you could try (that I have played around with) is using several high-power schottky diodes with a nice 8-bit adder and a digital potentiometer. That way you can adjust the voltage drop at relatively small intervals in much the same way you would with the op-amp setup. However, this one should be more robust about handing oscillations.
 

Hiemal

0
Joined
Dec 27, 2011
Messages
1,443
Points
63
Re: SuperBoost Drive

About your fully adjustable test load - with just a MOSFET, a resistor, an op-amp and some other parts, it won't work. I've tried it before, and unfortunately, there's something about the high-frequency nature of the IC that throws off the op-amp or something. I remember a while ago I posted about this, but it was probably over a year ago at this point.

One thing that you could try (that I have played around with) is using several high-power schottky diodes with a nice 8-bit adder and a digital potentiometer. That way you can adjust the voltage drop at relatively small intervals in much the same way you would with the op-amp setup. However, this one should be more robust about handing oscillations.

I'm wondering if maybe the addition of a low pass/high pass filter to the circuit would help calm the oscillations down. I know there's gotta be a way to fix it. Maybe another op amp and some cap/resistor magic or ...

Just something to fix it
 
Joined
Jan 14, 2011
Messages
3,816
Points
63
Re: SuperBoost Drive

That could work, but you would be dissipating a lot of power through that filter. You could potentially get away with a current mirror or a current averager so that you can generate the identical signal from a linear supply?
 

Hiemal

0
Joined
Dec 27, 2011
Messages
1,443
Points
63
Re: SuperBoost Drive

That could work, but you would be dissipating a lot of power through that filter. You could potentially get away with a current mirror or a current averager so that you can generate the identical signal from a linear supply?

It'd be a pretty high value filter so it shouldn't dissipate a lot of power. Nothing significant.
 
Joined
Jan 14, 2011
Messages
3,816
Points
63
Re: SuperBoost Drive

Oh, sure. I guess that works. But if you're shoving the driver through a filter... doesn't that defeat the purpose of the test load? To see the ripple, the flaws with the driver?
 

Hiemal

0
Joined
Dec 27, 2011
Messages
1,443
Points
63
Re: SuperBoost Drive

Oh, sure. I guess that works. But if you're shoving the driver through a filter... doesn't that defeat the purpose of the test load? To see the ripple, the flaws with the driver?

It wouldn't be through any part of the driver that actually takes the measurements and whatnot. The filter's purpose is solely to prevent the op amp from going nuts, and shouldn't interfere with the driver's operation if done correctly.
 




Top