AMC7124 driver chip

[12dstring]

Anyone interested in these at all? I got the idea from a thread from last year about the AMC7135 boards that DealExtreme sell.

The AMC7135 is a 350mA current regulator designed for LED driving, and has good reviews on here. It's >90% efficient and has a 0.1V dropout voltage. The only problem is 350mA is too high for lower power diodes, eg the blu-ray ones.

Reading the datasheet for the AMC7124 though, it has 3 seperate channels, configurable to output 20, 60 or 120mA - which can be paralleled to produce a total range of 20, 40, 60, 120, 180, 240 or 360mA. You could always parallel two seperate chips to get an even greater range, and with all the attractive features of the 7135; high efficiency, 0.1V dropout, 2.7-6V voltage range, very small...
They'd be great for PHR-803T, DVD-R and CD-R diodes for example.

I've tracked some down and have a quote for a minimum 50 of them for $1.16 (or about £0.75 in my money) each.
They're pretty small (3x4.9mm) so I also thought about getting some little pcbs made up which could fit inside an Aixiz housing.

Just wanted to find out if there's any interest in these first


Dave

 

[rkcstr]

I haven't used the amc7135 personally, but I've heard people say they can be tricky to use properly, specifically that they don't work well with an output cap with the protection diode (?). Also, though these are design for constant current, it is for LEDs and not laser diodes, so they should be tested for stability (no on/off spikes or transients during operation) with a scope. And ICs like this can be tricky to work with in general, like needing caps across various pins for stability and what not.

So, my point is basically that these may not work right out of the box, so I would probably consider their use as experimental (and needing proper testing), and not for immediate use with a laser diode.

 

[12dstring]

Thanks rkcstr,

You obviously know a lot more about drivers than I do! I'll see if I can order a few for myself at a reasonable price and test them out.

 

[Viracocha]

Intresting little device, you could probably put a resistor or small pot on the output to adjust the current to better suit your needs. However that would be an inefficient method and could produce a bit of heat.

As for the cost, if you bought those 1400mA units from DX you could get the chips for $0.44 (us) each, although they'll need a bit of unsoldering.

 

[12dstring]

Hi Viracocha,

The one that DX sell are the AMC7135, which are fixed 350mA regulators and as you say, you can get them to work inefficiently if you use a resistor to soak up some current.

The ones I'm talking about here are the AMC7124, which allow the output to be changed internally, so they'll be more efficient with lower power diodes.

I've got a quote for a smaller number, and the price jumps up to $2.80 a piece, which is a bit much for me. I'l have to think about it.

 

[12dstring]

I thought what the hell, so 50 of them are on their way from China.

Now I can post links, here's the datasheet if anyone's interested: http://www.addmtek.com/Datasheet/DD029-AMC712X_B(LF).pdf
and if anyone's interested in a few to try out then just let me know.

 

[12dstring]

They arrived today here's a photo of one next to an Aixiz module:

I don't have any diodes to test them on yet, but some are on the way. I'm also going to get some of these PCBs which will act as a little heatsink, and make them easier to work with.

This would be a typical application to power a PHR-803T at 120mA (~100mW):

You can also connect two more diodes to LED2 and LED3 to drive 3 at the same time if you wished.
A 5V voltage regulator (eg LM7805) can be used if you want to use a higher voltage source.

 

[phoenix3200]

Holy ***! Is that all the circuitry required?

Surely you must need something like a capacitor on the output to prevent voltage spikes to the LD...plus an extra diode for reverse voltage protection...

 

[12dstring]

The capacitor on the input protects from voltage spikes, it needs to protect the driver too as it can only handle 5.5V max as set up, so any voltage spikes are likely to damage that too. Having it on the input will also prevent the capacitor dumping all it's charge through the diode if it somehow disconnects and reconnects while still powered on, which would fry it.

A reverse voltage protection diode would be a good idea it's possible you might connect it the wrong way round. I left it out because I'd either build it with colour coded wires, or most likely I'd use a 5V regulator which have reverse voltage protection built in. If you do include one, you'll need to factor in the 0.7V drop it'll add.

This is the circuit I plan to use for testing:

It'll run a blu-ray diode from a minimum of 5.5V and reds from about 3V up. With a sufficient heatsink on the L4940V5, it'll take up to 30V input. With an overall voltage drop of <0.5V at 5.5V input, it's very efficient and so good for battery powered lasers. I'm also hoping it'll fit inside an Aixiz module, but that'll depend on how warm it gets.

 

[jonrobertd]

Hey sorry for the off topic ? but did you copper coat that Axiz or is it lighting?

 

[12dstring]

Just the lighting

 

[12dstring]

I got some empty PHR-803T sleds from Kendall at hightechdealz.com (thanks!) today. Here's my first test of the AMC7124 using one of the Red/IR diodes, driven at 40mA and 2.7V (flat CR2).

You'll also notice I'm not using a capacitor :-X that's because I'm naughty and I'm not at home at the moment so don't have any to hand. But it does show just how minimal the circuit can be.

The next step I'm hoping will be a PHR-803T diode, and a properly built circuit.

 

[jake21]

damn i want some

 

[jonrobertd]

You and me both! If that is all it would take to run reds (+a cap maybe) I could make a small pointer running on button cells(maybe?). And I asked about copper plate because I did mine and it looks just like that.