MR16 has the PT4105 which has a 18V max and 20V Absolute Max Vin. A bit intolerant of spikes for automotive duty but not that far off. It has a 25V input cap so that's fine. The problem is likely the inductor which looks way too small to provide an acceptable ripple at high step-down ratios, though there's no value listed on it.
Driver board list / regulator board list.
- Thread starter TorchBoy
- Start date
Help Support Candle Power Flashlight Forum
TorchBoy
Flashlight Enthusiast
Added: 8 mode 3.7-8.4 V. $7.35 for two-pack. Buck regulator based on Zetex C310 chip. Output 800-1000 mA on high. Modes 100%, 50%, 30%, 20%, 10%, SOS, strobe, fade in & out. No reverse polarity protection.
Just measured a DX sku.4382 board driving a Seoul P4. I'll try to post full data when I have a chance, but efficiency is actually pretty good. The circuit draws a between 1250 and 1400mA, which decreases as Vin increases.
Numbers are rounded slightly in some cases. I didn't feel like typing so many numbers in by hand. Yes, I am lazy.
Code:
Vin Iin Vled I led Eff
1.0V 1370mA 3.16V 290mA 67%
1.4 1390 3.27 440 74%
1.9 1330 3.38 575 77%
2.3 1270 3.41 685 80%
2.6 1240 3.45 765 82%
3.0 1250 3.53 870 82%Numbers are rounded slightly in some cases. I didn't feel like typing so many numbers in by hand. Yes, I am lazy.
Last edited:
Nope. No problems with any of my measurements. Based on my unscientific monitoring of the temperature of the circuit (I touched it, and it was running warmer at 1.0 Vin than at 3.0 Vin), I'd say that my numbers are fairly accurate. The thing is, since Iin is relatively constant at around 1.3A, you need to use this circuit with either rechargeables or D cells if using alkaline cells. I'll add the Iin and Vled data shortly.
Kaidomain added a new board today. As far as I know it isn't being discussed yet. It is simply listed as Boost and Buck circuit for XR-E Q5. No specs listed, but I just ordered some to play with.
http://kaidomain.com/ProductDetails.aspx?ProductId=5314
If any of the components look familiar to you or any other comments, let me know. Who knows, maybe we have a winner...
http://kaidomain.com/ProductDetails.aspx?ProductId=5314
If any of the components look familiar to you or any other comments, let me know. Who knows, maybe we have a winner...
The SO8 IC is a PIC (12F629), so it will have more than 1 mode.
I predict that efficiency will be poor on dimmed modes.
5kids
Newly Enlightened
Hey that's a Microchip part! I work for Microchip, but I don't have anything to do with chip design or programming...perhaps its time I get some "sample" parts.
TorchBoy
Flashlight Enthusiast
It looks like a work in progress. Hello? Can you read this? SHÖD I TEIP LRGIR OR SHOWT? At least someone spelled Cree correctly.
Has anyone tried these drivers from TCTEC?
http://www.emx.net.au/leddriver.htm
http://www.emx.net.au/leddriver.htm
gunga
Flashaholic
Hmm, is this new? I never noticed this one before. COuld be ideal for 1-2 AA type mods (with 14500 support). Anyone want to test this out? I may have to get one myself. I am interested in this!
I'd love to see some specs on this. Weird, lots of components on the bottom side tho...
kosPap
Flashlight Enthusiast
guys I have just completed some measurements of the KD flavor.
First of all it is a 3 groups-many modes board as following:
1. Low-Med-High
2. Low-Med-High-Strobe-SOS
3. Low-Med-High-Strobe-Warning Strobe (3 rapid blinks)-2Hz beacon-1Hz Beqacon-7 sec beacon-SOS
And here is what I measured:
So....this mode is going in a D26 module with 2AAs in a 3 cell Surefire, IF I find a way to lengthen the +contact spring!
First of all it is a 3 groups-many modes board as following:
1. Low-Med-High
2. Low-Med-High-Strobe-SOS
3. Low-Med-High-Strobe-Warning Strobe (3 rapid blinks)-2Hz beacon-1Hz Beqacon-7 sec beacon-SOS
And here is what I measured:
So....this mode is going in a D26 module with 2AAs in a 3 cell Surefire, IF I find a way to lengthen the +contact spring!
Last edited:
kosPap
Flashlight Enthusiast
Greetings all!
these are soem raw data from measurements of mine on 4 boards:
Dx 3256, KD 4982 Keenan (lattest version), unknown keenan variant (previous versiopn) and the driver pulled from a Derrelight 1SM-1 module. caution on the latter, it is a 6V maximum driver!!!!!
tables deleted
I am rather short on time so I will be updating thsi post with better tables and graphs within 2 days.
till then, Stay cool Kostas
these are soem raw data from measurements of mine on 4 boards:
Dx 3256, KD 4982 Keenan (lattest version), unknown keenan variant (previous versiopn) and the driver pulled from a Derrelight 1SM-1 module. caution on the latter, it is a 6V maximum driver!!!!!
tables deleted
I am rather short on time so I will be updating thsi post with better tables and graphs within 2 days.
till then, Stay cool Kostas
Last edited:
kosPap
Flashlight Enthusiast
Updates 11/30/08
i) I have inserted revised tables with a Lux per Watt inputed
ii) I have added some measurements of the driver of the Ultrafire C2 P4-bin HaIII flashlight
iii) I linked a table that compares the drivers hooked to 2xCR123 batts
well I have almost deactivated the above post. This is the new version of it.
Method and Details
All measurements (even current) were taken with a DMM. I did not get in the trouble of mmeasuring current with a high precision resistance 'cos my results are regular and right on the mark of the good drivers (Derrelight) specs.
I started with a number of AA MiMh batteries mainly for economy and easy duplication reasons. Somehow I was not satisfied and had a strange feeling so midway through the tests I added some CR123s combination. Maybe because I had noticed a trends for all the boards to fall out/in regulation somewhere between 3&4AA voltage.
My primary interest is 2xCR123 use and for just in case I added two old partly depleted Battery Station batts. Just to see how the driver performs in real life. Do notice how much these batts shag under load? (Tom a.k.a Silverfox how about if we chip-in some old ones for you to test?)
A General Observation.
MY lightbox is MY lightbox and I can only give you a reference. When I measure about 200 lumens I am around the 550 Lux figure. In this case though, you should expect 580+ 'cos measurements are made with a bare drop-in hence not accounting with flashlight body/switch resistance an glass losses.
First of all, I have a major observation/question with this board. When using lithium primary batteries efficiency changes while the input voltage and current draw are similar (3AA-2CR123 & 6AA & 3CR123 comparison). If my measurements are correct the performance with lithiums is Tasty!
My knowledge is limited and I do not know how to express this phenomenon but this driver seems linear in performance. The more you feed it the more it outputs.
Ok you get it! I was desperately lucky cos I kept feeding it more power beyond specs!
Anyway, this is a hottie! 1 Amp Straight.
Now why does we have the case od the DX3256 board in reverse? This driver gets less efficient when you feed it with NiMh…Go figure.
The only certain thing is that it is ment to be used with higher than 2xCR123 voltages.
Since versions of it can be found in most generic P60 drop-ins my measurements verify what I have experienced and heard from others. You buy these modules to get 300 lumens (200 realistically) that are achieved with 1amp of current. And you do not even get them cos with 2CR123s it outputs less….
This unknown version (more here) is close to the new one. All seems good
Anyway I am pleased with it 'cos having experienced a good amount of heat with it I thought it would waste much more power than it finally does. My 2 Surefires are now set with 3xCR123 so I AM getting 1amp after all!
Update:
Here are some tets results of the driver found in an Ultrafire C2 P4-bin HaIII flashlight
Pretty Impressive hah?
BTW here is a table that compares the five drivers when used with 2xCR123 batts. (well that was my need assembling a new C2 from parts) http://img390.imageshack.us/img390/953/drivercomparisonev4.jpg
That is all for now, Kostas
i) I have inserted revised tables with a Lux per Watt inputed
ii) I have added some measurements of the driver of the Ultrafire C2 P4-bin HaIII flashlight
iii) I linked a table that compares the drivers hooked to 2xCR123 batts
well I have almost deactivated the above post. This is the new version of it.
Method and Details
All measurements (even current) were taken with a DMM. I did not get in the trouble of mmeasuring current with a high precision resistance 'cos my results are regular and right on the mark of the good drivers (Derrelight) specs.
I started with a number of AA MiMh batteries mainly for economy and easy duplication reasons. Somehow I was not satisfied and had a strange feeling so midway through the tests I added some CR123s combination. Maybe because I had noticed a trends for all the boards to fall out/in regulation somewhere between 3&4AA voltage.
My primary interest is 2xCR123 use and for just in case I added two old partly depleted Battery Station batts. Just to see how the driver performs in real life. Do notice how much these batts shag under load? (Tom a.k.a Silverfox how about if we chip-in some old ones for you to test?)
A General Observation.
MY lightbox is MY lightbox and I can only give you a reference. When I measure about 200 lumens I am around the 550 Lux figure. In this case though, you should expect 580+ 'cos measurements are made with a bare drop-in hence not accounting with flashlight body/switch resistance an glass losses.
First of all, I have a major observation/question with this board. When using lithium primary batteries efficiency changes while the input voltage and current draw are similar (3AA-2CR123 & 6AA & 3CR123 comparison). If my measurements are correct the performance with lithiums is Tasty!
My knowledge is limited and I do not know how to express this phenomenon but this driver seems linear in performance. The more you feed it the more it outputs.
Ok you get it! I was desperately lucky cos I kept feeding it more power beyond specs!
Anyway, this is a hottie! 1 Amp Straight.
Now why does we have the case od the DX3256 board in reverse? This driver gets less efficient when you feed it with NiMh…Go figure.
The only certain thing is that it is ment to be used with higher than 2xCR123 voltages.
Since versions of it can be found in most generic P60 drop-ins my measurements verify what I have experienced and heard from others. You buy these modules to get 300 lumens (200 realistically) that are achieved with 1amp of current. And you do not even get them cos with 2CR123s it outputs less….
This unknown version (more here) is close to the new one. All seems good
Anyway I am pleased with it 'cos having experienced a good amount of heat with it I thought it would waste much more power than it finally does. My 2 Surefires are now set with 3xCR123 so I AM getting 1amp after all!
Update:
Here are some tets results of the driver found in an Ultrafire C2 P4-bin HaIII flashlight
Pretty Impressive hah?
BTW here is a table that compares the five drivers when used with 2xCR123 batts. (well that was my need assembling a new C2 from parts) http://img390.imageshack.us/img390/953/drivercomparisonev4.jpg
That is all for now, Kostas
Last edited:
guys I have just completed some measurements of the KD flavor.
First of all it is a 3 groups-many modes board as following:
1. Low-Med-High
2. Low-Med-High-Strobe-SOS
3. Low-Med-High-Strobe-Warning Strobe (3 rapid blinks)-2Hz beacon-1Hz Beqacon-7 sec beacon-SOS
And here is what I measured:
So....this mode is going in a D26 module with 2AAs in a 3 cell Surefire, IF I find a way to lengthen the +contact spring!
Is that the KD "Boost and Buck circuit for XR-E Q5" at KD i.e. sku.15880 at DX?
How does it compare with DX sku.7882 ?
kosPap
Flashlight Enthusiast
indeed it is...
BTW i was meaning to ask in due time.....is there a way to make this board a single mode????
well I have a suingle piece of 7882 just in case. So what I have done with it is put it in a ziploc wih a label only....sorry
BTW i was meaning to ask in due time.....is there a way to make this board a single mode????
well I have a suingle piece of 7882 just in case. So what I have done with it is put it in a ziploc wih a label only....sorry
