High Power LED Driver, 9 amps

Someone asked me to design a simple high power driver for the SST-90 LED.

UPDATED 12-18-09
All parts are in house for 2 prototype builds, boards are on order, should be here by 12-22-09.
I will be on vacation starting the night of 12-22 through the holidays, so the first build wont be until maybe the last week of Dec.
The following are specs and options for this driver.
Size
The board is 1.2" in diameter by about 0.5" thick. I will fit in a "D" size Mag light.

The voltage input range is 6 volts to 24 volts. The board will not operate any lower than 6 volts because of gate drive issues with the mosfet.
The input is NOT reverse protected, care must be taken during hook up. This is mainly due to size and efficiency issues.
The input to the board is ESD and transient protected. The protection starts to conduct at about 28.9 volts, if you apply this it will damage that part.
The board it self is ESD sensitive, proper ESD handling procedures must be followed. Most LEDs are ESD sensitive also.

Low voltage drop out setting.
The board can be configured at time of order to turn off at a certain voltage. This is to protect the board and the batteries.
The default voltage that will be shipped is 10 volts.
No charge options for the voltage drop out are 6v, 10v, 11v, 12v, 13.1v, and 14v.
You must apply about 0.5 volts more than the drop out voltage for the driver to turn on.
Once on it will turn off when the input drops to the set voltage.
Other voltages can be custom set but may incur an additional charge for me to order more parts.

Output
This is a true constant current driver.
The user has some options when ordering for the current set points.
There are 2 connections on the board, shorting either to ground will select one of the current settings.
The user can order the following current settings 5, 6, 7, 8, 9, 10 amps on either selection.
So you can have 2 different drive currents.

The user could also hookup a user supplied 100K pot for linear dimming. You must specify you want to do this
at time of order and you can not use the switch method as above.

When ordering you must specify the forward voltage of the LED you plan to drive.
The design forward voltage is between 2 and 4 volts. If you want to work outside that range I can adjust the driver to do so on a special bases.
Driving voltages outside of the designed range could damage the board.

The board has output open protection if the LED where to open or become disconnected.

Should be able to start shipping sometime in January.
The cost will be $50

As lights are generally cylindrical, most people like/expect a round driver, but more importantly the biggest single market I suspect will be Mag conversions.
Can you confirm that it would fit inside a D-Cell Maglight tube?
It would be preferable, from a packaging point of view, to fit against the back of the heatsink rather than being lengthways.
The HipFlex is 1.4" round and just a bit too big.

ps I like the format that allows a low current trigger to be used rather than pushing full power through the switch.

Nick

Is this a true Constant Current driver?
If so I need at least one, and if you can make them fit in a D-cel mag I am going to need more!!

I think there will be a huge market for a true CC driver wich can deliver 10A and can handle at least 20V input.

Greetz Johan

P.s. Have you got some efficiency numbers for a 14,8V supply and 1 SST-90 @ 9A?

I would be in if it would fit in a D Maglite and if the price would be lower .

rayman

rayman said:

I would be in if it would fit in a D Maglite and if the price would be lower .

rayman

Click to expand...

I think that US$50,- is a fair price, it is only €33,- just a bit more than Der Wichtels driver. But it is capable of 10A

I'm in for at least one board

D Mag size would be great

What kind of time frame are we looking at before we can actually get
our hands on them ?

These would be awesome for automotive applications. how many SST-90's could one of these drive?

I'd be in for one too

Doh!Nut said:

As lights are generally cylindrical, most people like/expect a round driver, but more importantly the biggest single market I suspect will be Mag conversions.
Can you confirm that it would fit inside a D-Cell Maglight tube?
It would be preferable, from a packaging point of view, to fit against the back of the heatsink rather than being lengthways.
The HipFlex is 1.4" round and just a bit too big.

ps I like the format that allows a low current trigger to be used rather than pushing full power through the switch.

Nick

Click to expand...

When it comes to prototyping drivers...
Build and perfect your circuit on a breakout board first, when you have all the components right then figure out a way to stuff them into a round volume.

theres enough space in a mag D body to stack 2 trimmed boards, that should be plenty of room for those "bigger" components

Ok to answer the questions:

This is a true current source. It is a buck topology converter. There will be some ripple current but not much and it will be at a very high frequency. To make a converter handle this much current requires some larger components. All these components tend to be square in shape. So a round board is a little harder. I did some quick layout options and the smallest diameter I can go it 1.2" maybe a little bigger. I will work on that and let everyone know. Board coverage is already VERY high.

Efficiency should be somewhere between 80-90%. Its hard to calculate and until I build some and measure it that is as close as I can get. But I sacrificed a little efficiency for size.

I don't think I can get the price any lower. Because of the high current the cost of components goes up. Also building low quantities drives the prices higher and I don't plan on the demand being high enough to justify true production runs.

I think with some component changes we could drive at least 3 SST-90s in series.

Time frame has now been pushed out by needing a new board layout

I will work on the new layout and let you know. The design is finished. Boards I get from a prototype house and take about 3 days to turn. All components are in stock at various suppliers and should take about a week to get here. Then I need to build and verify the design. Everything said and done we are most likely looking some time in Jan 2010.

I don't have any of these LEDs for testing. I'm just going to drive the power into a few power diodes I have that can generate the same Vf at that current for testing. There cheaper if I blow them up.

Illum said:

When it comes to prototyping drivers...
Build and perfect your circuit on a breakout board first, when you have all the components right then figure out a way to stuff them into a round volume.

theres enough space in a mag D body to stack 2 trimmed boards, that should be plenty of room for those "bigger" components

Click to expand...

Because this is a higher power high frequency drive, bread boarding is out of the question. Bread boards and wire wrap just create issues in the design. Even brass boards will have parasitic and power issues. The easiest way to prototype these designs is to just get a run of boards made so you can measure what the final product will preform. That and all the parts only come in SMT.

2 boards would not help either. When designing high current and frequency switching drives you need all the components as close together as possible. Spiting up the components on separate boards is going to cause tons of EMF, parasitic, and transient issues. The size of the board is being driven by 4 main components. I will try to make the smaller board but heat maybe an issue then as well. The board is used to dissipate heat. The smaller the board the less area you have to shed the heat.

This is good news!
I would be in for one to drive 3 SST-90 in series...

I'm in for 4

Ok, I was able to get the board to 1.2" in diameter. That is slightly smaller then a "D" size battery (1.35"). However for the smaller size I had to give up some things.

I will need to check the heat dissipation of the board when I build it. I had to reduce the size of the power plain which will reduce cooling surface area.

There are only 3 main connections, + power, -power, and + LED. The ground to the LED will need to be connected in your wiring. There was no room on the board left for another hole. There is a 4th smaller connection for a switch. When this connection is pulled to ground the drive will turn off. However there will still be some current draw. The chip specs say 100 uA and the 10K ohm load through the switch. Also there are leakage currents through diodes, caps, and other parts.

Finally, there are no mounting holes. The board will need to be attached to something with epoxy, RTV, double sided tap on the bottom, ect. You should be able to encapsolate the whole board if you want provided you use a thermal conductive epoxy.

The price stands at $50 plus shipping. I think there is enough potential demand to warrant going forward. I should place orders for boards and components by the end of this week. Then build up and test 2 units. If the tests come out good I will take some orders and ship.

When you order you must specifiy the input voltage range you plan to use, the current drive you want, and the planned Vf of the LED. You also need to specifiy the Minimum operating voltage you would like. This will turn the drive off when the batteries get low to protect the cells. I need this information so I can tweek the control points and make the drive most effcient.

Please feel free to post any comments. More information to follow.

I have an SST-90 here. If you want me to test one of your boards for you send me a PM.

Mac

See I told you there would demand for this driver!

I get the 1st two boards right?

Is there any way to make this a multi mode board?
Perhaps by using a already available off board board?

Does it have protected output? EG if there is no load will it fry?

Can you leave the enable pin connected and just switch the Vin to it?

Packhorse said:

See I told you there would demand for this driver!

I get the 1st two boards right?

Is there any way to make this a multi mode board?
Perhaps by using a already available off board board?

Does it have protected output? EG if there is no load will it fry?

Can you leave the enable pin connected and just switch the Vin to it?

Click to expand...

Yes sir, you get dibs after verification. :buddies:

You can PWM the disable line for true color dimming if you would like. I was going to test it to see how it works. If I used a pic chip I could do what ever you guys want. Which I may do in the furture, but for now I'm using a standard controller. I don't see a problem with just leaving the disable line not connected and switching the Vin. That is something I will test to make sure. The transient protection should snub out any unwanted startup ringing. An open output should be just fine. The driver has fly back protection and all items are rated for the input voltage.

I'm really tied up working on a high wattage DC HID ballast that has been a pet project for 2 years now. So it would be a long time before I get a pic chip version.

Subscribed.

It really needs to fit a Mag D Tube

It would probably be O.K. if you had to split it into 2 boards, as long as the components that need heatsinked are all on the same board so it could be mounted to the LED heatsink.

Watching this one for updates!!

Read my previous posts.
I was able to get the board down to 1.2" in diameter.
A "D" cell battery is 1.35" in diameter. It should fit now.