OPEN SOURCE: "CC-Boost" - 2.4 Amp boost driver - RC1

[rhd]

There is still a chance it could be
something else, though, but at this point
we have prettymuch narrowed it down to
either the chip or one (or more) of the
resistors. I want this thing to work!

To be fair, "this thing" does work.

Just not when the two of you make them

 

[crazyspaz]

I got mine from DigiKey and had the same
results. My mind keeps going back to the
IC. Either something is causing them to
fail or they were bad in the first place.

Did you try the chips I sent you? If so, results?

 

[IWIRE]

To be fair, "this thing" does work.

Just not when the two of you make them

That was why I built one and took pics of it working. People were making it sound like they weren't sure the drive even worked. Maybe mine works because I know nothing so I'm extra careful when assembling them.
Good luck with the drives. Cya !
Intelligence isn't thinking you know everything. It's questioning everything you think you know.

 

[ARG]

2.1A results for the buck driver. Apologies for screen photo. Couldn't find my thumb drive.

Got to wait before doing the 10A test, the test load heats up fast

Edit: Can't get it above ~2.4A at the output with the pot, do I need to change a set resistor?

 

[rhd]

2.1A results for the buck driver. Apologies for screen photo. Couldn't find my thumb drive.

Got to wait before doing the 10A test, the test load heats up fast

Edit: Can't get it above ~2.4A at the output with the pot, do I need to change a set resistor?

Nope, the pot works up to 10A.

I think it was set to 4 or 5A when I mailed it. What are you powering it from?

 

[ARG]

Edit: Nevermind Measuring current at the input. Whoops.

 

[ARG]

9A (didn't want to run it too close to the 10A limit, but I can go higher if you want)

5A

2.6A

I tested it at more currents, but only took photos at those ones. It looks like the start up spike is independent of the output current.

Edit:
Here's the output with a 220uF electrolytic directly across the driver output. I also tested it with a 4.7uF cap, but noticed little difference.

Edit 2:
I can grab some large caps from a school lab and test it with a higher capacitance later. I don't know a ton about driver design, but it looks like it would benefit greatly from a large output capacitor. It may be best to expand the foot print a little and add in a 2917 size 470uF tantalum capacitor. It's costly ($1-$2 per tant) and would expand the driver footprint quite a bit, but I don't think anyone expects a 10A buck driver to be tiny or cheap

 

[crazyspaz]

10A buck? Is this something we will see open source, or at least be able to buy?
(Although lets be honest, its evident i'm shite at making drivers anyway )

 

[IWIRE]

(Although lets be honest, its evident i'm shite at making drivers anyway )

Not for long. You will have some new pieces in about 3 days. :na:

OK. If anybody is still watching this channel. I played with CC Boost some more. I got out my stopwatch and my IR temp gun. I took the drive and embedded the inductor in a blob of thermal adhesive on a heatsink and fired it up. After two minutes the current started dropping off. The hottest component was the ZXCT at 40C. I heatsinked the ZXCT and the run time increased to 3 minutes before current started dropping off. I removed the heatsink from the ZXCT and installed a heatsink on the IC. After running 4.5 minutes I had to shut it down, my testload was getting pretty hot, but the drive was still holding current steady. The temp of the IC was 34C and the temp of the ZXCT was back at 40C.

 

[djQUAN]

ARG, which circuit revision is that driver? Is there a complete schematic somewhere? I might be able to figure out where the spike is coming from.

 

[ARG]

ARG, which circuit revision is that driver? Is there a complete schematic somewhere? I might be able to figure out where the spike is coming from.

It's not the CC boost, it just got thrown in with this thread Should probably have its own if there's going to be a discussion on it, to keep it from becoming confusing as to which driver is being talked about.

The first post on it is here: http://laserpointerforums.com/f67/o...mp-boost-driver-rc1-89556-15.html#post1325406

Doesn't look like it's intended to be open source. RHD will have to comment on this.

 

[rhd]

10A buck? Is this something we will see open source, or at least be able to buy?
(Although lets be honest, its evident i'm shite at making drivers anyway )

No plans re: the driver's future at the moment. I have a 25A version that I want to test, as well as a 5A version that's a bit smaller than a flexdrive. But these guys are special to me and have taken a lot of work to get to where they are. Anyway, I've chatted about this topic earlier in the thread somewhere.

ARG, which circuit revision is that driver? Is there a complete schematic somewhere? I might be able to figure out where the spike is coming from.

My bad for the confusion. That's a different driver, a buck, that I sent ARG.

On the bright side, that spike is no mystery. It's output prior to the current sense IC coming online. It's an easy fix with extra capacitance after the sense resistor rather than before. It happens because the inductor capacitor holds the voltage below what the IC needs to boot up, and then charges up a bit too high before the IC comes online. You just put a cap after the IC and it eats the startup spike.

Not for long. You will have some new pieces in about 3 days. :na:

OK. If anybody is still watching this channel. I played with CC Boost some more. I got out my stopwatch and my IR temp gun. I took the drive and embedded the inductor in a blob of thermal adhesive on a heatsink and fired it up. After two minutes the current started dropping off. The hottest component was the ZXCT at 40C. I heatsinked the ZXCT and the run time increased to 3 minutes before current started dropping off. I removed the heatsink from the ZXCT and installed a heatsink on the IC. After running 4.5 minutes I had to shut it down, my testload was getting pretty hot, but the drive was still holding current steady. The temp of the IC was 34C and the temp of the ZXCT was back at 40C.

Neat! And unexpected. I wonder if it's actually the ZXCT heating or just its proximity to the sense resistor. Did it seem to heat more than the resistor.

Anyway good news on the testing. Thanks for vindicating the driver, and confirming that I'm not the only one who can reflow it!

 

[rhd]

It's not the CC boost, it just got thrown in with this thread Should probably have its own if there's going to be a discussion on it, to keep it from becoming confusing as to which driver is being talked about.

The first post on it is here: http://laserpointerforums.com/f67/o...mp-boost-driver-rc1-89556-15.html#post1325406

Doesn't look like it's intended to be open source. RHD will have to comment on this.

Thanks so much for testing it truly appreciate it.

I should fork it out into its own thread at some stage. I'm just not sure what, if anything, I want to do with it.

I'm a big fan of open source, but I'm not sure the model quite works for drivers.

I prefer the concept of working closely with a few other people, rather than tossing a driver into the wild where 80% of people are focussed just on cutting driver costs, rather than tweaking a design to perfection. It's the collaboration that I enjoy.

I certainly have zero intention to produce or sell drivers.

I just haven't figured out what I DO want to do with them yet.

I've got a lot of really good buck drivers, and should probably figure out something. But Lazeerer has the board well served with awesome bucks, so there's not really any urgency to doing anything on the open buck front, IMO.

 

[crazyspaz]

Not for long. You will have some new pieces in about 3 days. :na:

Again, thanks a million!

No plans re: the driver's future at the moment. I have a 25A version that I want to test, as well as a 5A version that's a bit smaller than a flexdrive. But these guys are special to me and have taken a lot of work to get to where they are. Anyway, I've chatted about this topic earlier in the thread somewhere.

Ah yes, I remember now. If you ever have any spare, just know you won't have any trouble finding people to sell them to

Anyway good news on the testing. Thanks for vindicating the driver, and confirming that I'm not the only one who can reflow it!

>.>
<.<

I prefer the concept of working closely with a few other people, rather than tossing a driver into the wild where 80% of people are focussed just on cutting driver costs, rather than tweaking a design to perfection. It's the collaboration that I enjoy.

I wish I could have helped, but there's really nothing I could have helped with. I can tell you however, that I was more attracted to the idea of making drivers myself (seemed like a pretty cool and foreign idea to me) and being able to give them to people for next to nothing to play with rather than just trying to save myself a couple bucks. I have always used Lazeerers drives and will continue to do so. In fact, I need to get some more from him soon for somethings...

I've got a lot of really good buck drivers, and should probably figure out something. But Lazeerer has the board well served with awesome bucks, so there's not really any urgency to doing anything on the open buck front, IMO.

I think we went over this earlier in the thread when these were brought up, and once again I agree with you. Not to say releasing it as open source would run Angelos out of business (seeing as very few people are interested in making drivers, it seems), but it certainly doesn't seem like a very cool thing to do.

 

[ARG]

On the bright side, that spike is no mystery. It's output prior to the current sense IC coming online. It's an easy fix with extra capacitance after the sense resistor rather than before. It happens because the inductor capacitor holds the voltage below what the IC needs to boot up, and then charges up a bit too high before the IC comes online. You just put a cap after the IC and it eats the startup spike.

Glad to hear that the problem is an easy fix

Let me know if you want me to remove/add some caps on the board to test values before you fab the next revision.

Thanks so much for testing it truly appreciate it.

Happy to help, thanks for the test load!

 

[djQUAN]

Glad the problem has been figured out

I prefer the concept of working closely with a few other people, rather than tossing a driver into the wild where 80% of people are focussed just on cutting driver costs, rather than tweaking a design to perfection. It's the collaboration that I enjoy.

I would love to join and contribute but I'm too far that shipping them to me for testing is a no go