High efficient 5A/9A Buck Converter for SST-50, SST-90

[Der Wichtel]

The wait is finally over.

There are two driver versions: 5A and 9A (please specify while ordering).

Here are the specs:

- High efficiency ( 94% - 98%)
- Input voltage range: 6V-22V
- Output current: fixed to 5A or 9A
- The input voltage must be at least 2V higher than the output voltage, otherwise output current will get lower.
- 1mm² silicone wires preinstalled
- Diameter: 26mm
- Height: 13mm

The driver must be potted to the heatsink, at least if output power goes over a single SST-90.

More information can be found in the datasheet:
Datasheet
Please read it closely until the last page

and as always please contact me via pm or email for order information.


Price:
standard version 35 Euro
adjustable version 41 Euro

Shipping:
( only has to be paid once each order)
- Germany 2.20 Euro
- International 3.45 Euro

- optionally registered mail 2.05€

Shipping can take a little bit longer because soldering these drivers became much harder

[Icarus]

I'll take one (5A).

[SmurfTacular]

NICE!!!! Can't wait to get a hold of one! I'll definitely buy one when I get my funds together.
I have a few questions if you don't mind.

Will these be available with multiple modes? Or even (as donn mentioned) a potentiometer/dimmer?

What gauge of wire would you recommend? is 18 gauge OK?

How exactly do you mount the driver onto the heatsink?

[Drywolf]

I'll take two 9A and one 5A model.

[donn_]

Will these be available with multiple modes?

Or can they be used in conjunction with one of George's dimmers?

[Der Wichtel]

Hi,
unfortunately it is not possible to dim the output.
The regulation is a little bit complex so that simple pwm won't work.

the dimming part on this one will be as difficult to realize as the driver part. I'm working on it and I already have two ideas of how to do it.

One simple way and one much more difficult. The circuit for the simple way is already on the driver but I don't think it will work that easy. Well only a test will clarify which way will work.

The more difficult way will be able to use a potentiometer AND a microcontroller.

Good night

EDIT: 18 gauge / 1mm heat resistant silicon wires are preinstalled. Potting can be done with Arctic Alumina for example.
I'm using epoxy mixed with 65% aluminiumoxid which works very good for this. Besides of that it is cheap

[Mettee]

just saw this, super cool!

[TorchBoy]

Here are the specs:

- High efficiency ( 90% - 98%)
- Input voltage range: 7V-20V
- Output current: fixed to 5A or 9A

...

More information can be found in the datasheet:
Datasheet ( let me know if there is something unclear or somthing I should add )

The datasheet appears to indicate 6 V to 20 V and 5.6 A or 8.6 A. Which is it?

[Der Wichtel]

The datasheet appears to indicate 6 V to 20 V and 5.6 A or 8.6 A. Which is it?

It should be 6-20V.

The efficiency tests were done with 5.5A and 8.6A because I did not have the correct current setting resistors during the test. However the results are similar.

Just saw that you have a list of drivers. Nice work!
So the correct values for your table are:

Vin: 6-20V
Number of LEDs: 1-4
Efficiency: 90%-98%
Output Current: 5A / 9A
Number of Modes: 1
dimension: 26mm diameter, 13mm high

[TorchBoy]

So the correct values for your table are: ...

Right, done. Good work, BTW - the 9 A model in particular fills the gap at the top end very nicely.

[FlaCracker]

43.05 Euro sent for:

9A Buck Converter: 35 Euro
Outside Europe: 6 Euro
additionally registered mail: 2.05 Euro

[FlaCracker]

Does the driver shut off below 6v?

[Der Wichtel]

The cut off voltage is 4.5V.
If the voltage is below 6V output current will decrease slightly

[lucca brassi]

Work of Art ...... Respect!!

How you done with heat manegment ?....max temp?

Some potting in the Al or Cu capsule will maybe reduce temp.

You use PWM I think or current walley on uC ?

Regards!

[Der Wichtel]

How you done with heat manegment ?....max temp?

Some potting in the Al or Cu capsule will maybe reduce temp.

yes, as described in the first post.

You use PWM I think or current walley on uC ?

Regards!

no modes yet

[Nenth]

for 1 9A

Cheers

[Mdemon]

Hi, I'm interested in driving SST-90s with this and I'm using a 10x10Ah 1.2v D cell NiMH battery pack. As I don't want to fry the battery pack (nor the SST-90s), what configuration would you recommend as being "safe" please?

Can I drive three or four SST-90s at the same time or just one? What would be best with your driver please? (I'd go 5A or 9A on your recommendation)

Sorry for the Noob question but I don't want a moment!

[irv_usc]

Hi, I'm interested in driving SST-90s with this and I'm using a 10x10Ah 1.2v D cell NiMH battery pack. As I don't want to fry the battery pack (nor the SST-90s), what configuration would you recommend as being "safe" please?

Can I drive three or four SST-90s at the same time or just one? What would be best with your driver please? (I'd go 5A or 9A on your recommendation)

Sorry for the Noob question but I don't want a moment!

that's an easy one. you would wire the LEDs in series. as long as your battery pack is more than 2V higher than the Vf total of your LEDs it will run at 5A or 9A. For simplicity if your LEDs were 3.5V each, you could only run AT MOST 3 LEDs with your battery pack, assuming the cells in your pack are in series for 12V.

[Mdemon]

Thank you. Is it as simple as that? 3 x 3.5 = 10.5v + whatever else is needed. That would mean that the thing would be out of regulation pretty quick wouldn't it? How long would that be?

Would 2 x SST-90s mean I could use around half of the battery power "within regulation"? Math not my strong point...

[Der Wichtel]

Yes, it will be out of regulation very fast and the output will get lower.
Furthermore there are voltage drops at the switch and the cables.

With 12V I wouldn't use more than 2x SST-90. But with one or two more NiMh batteries you can think of powering up 3 LEDs. However keep in mind that your batteries must be able to handle the current

[rantanplan]

Got my 5A driver yesterday ... it works just perfectly. I used it to replace the faulty driver of a Trustfire ST-50 light. Thinking about the cooling of the driver took me a while. The hint using aluminiumoxid mixed with epoxy is probably good, but I had no clue where to get aluminiumoxid quickly without a couple of days "delay" . So I used the classic approach: Using (lots) of arctic alumina to connect the important parts of the PCB to an aluminiun piece which itself is glued to the light engine. Should be fine with approx. 1.2W loss at the driver. Powered by 2x protected 18650 battery current starts just below 3A, so I´m pretty sure that the SST-50 gets its full 5 amps.

Now I got a fully powered SST-50 light, which is now the brightest light in my collection, but I doubt that it will be more than a seldom used demo light. Too much heat for the small body and compared to a Mag host with 4x XR-E, the lumen advantage of the SST-50 is barely visible ... maybe I should order a top bin SST-50 from Fred or even jump onto the SST-90 train

Thanks Yitao for this nice driver!

One small suggestion is that you could add the pin positions to the datasheet, or maybe the color code of the wires too. A small sketch of the PCB with the pin locations marked would be helpful. V+ and GND markings are easy to find on the PCB, aside from clear wire colors. But on the output side it took me a while to find the L+ mark and I couldn´t find a L- ... color wires are helpful, but I heard rumors that there are lots of color-blind people out there

[Der Wichtel]

Nice to hear that it worked fine.

The writing on the pcb is not very good indeed. I think I chose a too small size for it.

I will change it in the next update, thanks for the info.

[flashfiend]

rantanplan, jump on the '90 train. Even at 5A it should blow you away.

[Drywolf]

I'll take two 9A and one 5A model.

Got mine today!
Thanks.

[Drywolf]

I'm using epoxy mixed with 65% aluminiumoxid which works very good for this.

It's the left pic below which must be potted closest to the heatsink, right? Should I pot the entire assembly?
Thanks,
Frank

[Stereodude]

DW, I believe you have a mistake in your datasheet. In the efficiency graph on the top of the 2nd page the red line is labeled 13.5V @ 5.6A, but it shows input voltages all the way down to 10V. Since, Vin must be greater Vout on a buck converter, something isn't quite right there.

[aurum]

no ... vin is 10-20V, vout is lower (there is no graph for Vout). Iout is 5.6A ... 13.5v is just an example to calculate the generated heat (13.5*5.6=75.6W Pin / -> Pin*efficiency=heat in W 75.6W*0.02=1.5W loss in the driver) I hope I'm right

[Stereodude]

no ... vin is 10-20V, vout is lower (there is no graph for Vout). Iout is 5.6A ... 13.5v is just an example to calculate the generated heat (13.5*5.6=75.6W Pin / -> Pin*efficiency=heat in W 75.6W*0.02=1.5W loss in the driver) I hope I'm right

You can't plot efficiency unless you fix the output condition and vary the input voltage (or fix the input and vary the output). Given those graphs there must be a fixed output condition as the input is changed. For example driving 3 SST-50's in series at 5A would be the fixed output condition. You would then vary the input voltage, record the data, and create an efficiency plot.

It seems that the output condition for the red trace is not 5.6A @ 13.5V into a dummy load.

[Kaulquappe]

Hi, i'm trying to decode the Efficiency vs. Voltage diagram. In the Y-axis there is the efficiency in % and in the Y-axis the Vin Voltage in Volt.

But why are there several lines and not only one?

Regards from Switzerland

Luca

[aurum]

I think it's:

first line: 1*SST-50/SST-90 @ 5.6A
second: 3*SST-50/SST-90 @ 5.6A
third: 1*SST-90@ 8.6A
fourth: 3*SST-90@ 8.aA