hipFlex

LEDBE
trout said:
Fast work George
I just had a read on Taskled .com and you say to keep the switch wires short ie 4 inch .

Are there any issues with the length of the wires to the leds
Like if the HipFlex was in the battery pack and powering 2 light heads on the bars so maybe 18 inches long
Click to expand...

hipFlex has the onboard pullup resistor like the bFlex V2 and maxFlex3. I'm just being my usually conservative engineering self when I say 4". It should be fine with longer wires - but I'm guaranteeing 4"...

cheers,
george.
 
georges80 said:
hipFlex has the onboard pullup resistor like the bFlex V2 and maxFlex3. I'm just being my usually conservative engineering self when I say 4". It should be fine with longer wires - but I'm guaranteeing 4"...

cheers,
george.
Click to expand...


Sorry George , I was meaning the wires to supply the leds from the HipFlex If there was any problem them being long
 
That's awesome George. For each table I assume your numbers are the max setting for that particular table, what do the steps look like, ie At the 1000ma max setting are the others 750, 500, 250 etc...?

No idea how low you want to go but at 2800 it would be cool to have a ~400ma setting so when you turn it on people are like that's 'cute' ( @ 150lm per MC-E, i will have 4 Each in P wired in S ) but then when you go to full power wow! Is there a way to use a momentary to go to full power no matter what setting you are currently at?

Is it possible to program our own tables, ie can we reprogram your firmware? I am going to be heat limited in my installation so would like to figure out exactly how high i can run and make that the limit and step down from there.

I am sure all of these have been asked at some point so sorry for the redundancy.

Cant wait!
Jason
 
Having a moon mode like the maxflex would be a plus for me. It came in handy for long duration light during the last hurricanes.
 
Current tables are fixed with a choice of 4
1000mA max
1400mA max
2000mA max
2800mA max

The low on all tables is effectively around 50mA (long runtime). The other levels are essentially current halving from max.

e.g. the 2000mA table on the 5 level UI's (UIP and UIB) would be approximately

50mA, 250mA, 500mA, 1000mA, 2000mA

No, I don't plan to offer user configurable current tables. Way too much work and would more than likely become a support nightmare when folk mess things up and don't know why.

I have to answer enough emails with folk that forget to set or clear some option.

cheers,
george.
 
Ah yes, had not considered your support nightmare letting us mess with the software plus the current steps you have sound perfect anyway.

Jason
 
georges80 said:
The low on all tables is effectively around 50mA (long runtime). The .
Click to expand...

I foresee a problem with this but I may be wrong.

50mA is only 12.5mA on each die and I think this is below the threshold on the graph where the lumens/W start to drop off rapidly at the low end. Would it not be better to set a low level of around 80mA?(which puts 20mA on each die) and hits the graph round about the peak efficiency for the LED. I can't remember which thread it was on but the peak hit around 150 lumins/W at around 20mA for the newer R2 dice.

Also there will be significant color shifting at too low a drive current.

Marco
 
marcopolo said:
I foresee a problem with this but I may be wrong.

50mA is only 12.5mA on each die and I think this is below the threshold on the graph where the lumens/W start to drop off rapidly at the low end. Would it not be better to set a low level of around 80mA?(which puts 20mA on each die) and hits the graph round about the peak efficiency for the LED. I can't remember which thread it was on but the peak hit around 150 lumins/W at around 20mA for the newer R2 dice.

Also there will be significant color shifting at too low a drive current.

Marco
Click to expand...

Um... I have actually been running hipFlex in the ~50mA configuration and it works great with the P7.

Oh - maybe because for current less than 500mA I'm actually using PWM with 500mA pulses to maintain 'tint' etc... I also have to do the PWM with 500mA pulses to maintain accurate/repeatable regulation through the buck converter circuitry. So, 50mA is actually 1:10 duty cycle 500mA pulses to the LED (at high frequency).

Hmm - maybe I did something right? 🙂

Besides, the whole point of 50mA drive isn't lumens/watt or tint etc - it is having long runtime to repair a flat tyre or set up camp or to 'limp home' or some other emergency situation.

So, in summary, don't panic :wave:

cheers,
george.
 
Hi George -you ain't no dummy 🙂
Can you go to a lower current than the 50mA?
I remember you said something gets unstable at lower currents with maxflex/bflex.

Or do the maxflex/bflex operate the same ,with the duty cycle?

Cheers
Dom
 
dom said:
Hi George -you ain't no dummy 🙂
Can you go to a lower current than the 50mA?
I remember you said something gets unstable at lower currents with maxflex/bflex.

Or do the maxflex/bflex operate the same ,with the duty cycle?

Cheers
Dom
Click to expand...

50mA is a realistic & reliable low for the hipFlex. On my tests it is a reasonably low level for 3 P7 LEDs (not stupid bright) and even more applicable to a single P7. Remember, that's effectively only 50/4 ~ 12mA per die (average).

maxFlex/bFlex are fully current regulated at the low end (not pwm). Just the nature of the different drivers and the current sense accuracy.

One way to look at it is that on hipFlex the current sense resistors are 3 parallel 0.2 ohm resistors. That means 0.2/3 = 0.0667 ohms.

So, if we have say 50mA, that means 50mA x 0.0677 = 3.3mV that would need to be sensed (accurately) for current regulation. That is below the noise level of the current regulation system and would not work. By using 500mA I'm up at 33mV sensed and I then PWM 1:10 to get an average 50mA to the LED(s).

In summary, what I'm blabbering on about is that achieving the low end of the current scale is not easy (and not have it flicker or vary with battery/load etc) and requires some algorithmic 'tuning' to get it stable and repeatable over 100's of units. Each driver has its own issue to achieve the low end of the current output and there are lots of trade offs that depend on how small a duty cycle, or how fast a PWM frequency or how low a sensed voltage works, i.e. there is some method to the madness 🙂

cheers,
george.
 
RalphStirling said:
George,

It would be nice to have even higher current modes for
driving these:

http://www.luminus.com/content1462

They will put out 1700 lumens at 13 amps, or 500 lumens at
3.2A. I'm itching to try one when they are released to
distribution.

-- Ralph
Click to expand...

Yes, many things would be nice 🙂

We are talking 13.5A at 3.8V => 51W and 1700 Lumens. That seems a pretty poor lumens/watt rating... I'd rather go with a 3 x P7 or 3 x MC-E and get much better efficiency...

A P7 claims 90 lumens/watt. That means at 2.8A (3.3V) = 9Watts we have 9 x 90 = 810 lumens. Can't see why you are so excited with the luminus part ???

hipFlex is optimized for 2.8A max output. Going to higher current would require much larger inductor, switcher IC, FET, polarity protection circuitry, sense resistors, heatsinking, PCB and on and on and on....

cheers,
george.
 
Ciao Giorgio,

your drivers are the best in the solar system...from what i know! :twothumbs
 
No above 2.8A(3.2A?) drive current like 1.2A in your Maxflex George? I would love to have an overdrive mode for a non handheld setup. why 4 current table not 5 like Maxflex?

thanks
 
rizky_p said:
No above 2.8A(3.2A?) drive current like 1.2A in your Maxflex George? I would love to have an overdrive mode for a non handheld setup. why 4 current table not 5 like Maxflex?

thanks
Click to expand...

2.8A max - that is the design spec. The switcher IC/FET is rated to 3A and I'm setting 2.8A as a conservative limit. You really think you can see the difference between 2.8A to a P7 versus 3.2A - bet you can't...

4 Tables since they seem like reasonable options for various P7/MC-E combinations. At most I may add a 5th table with 2.4A for folk that want a bit more runtime & less heat versus 2.8A for only a small drop in visually perceived light output.

Boards are in assembly, just waiting on one last IC to be shipped.

cheers,
george.
 
I'd agree with George on the max current. I have an MC-E hooked to a bench supply and can see no difference between 2.8A and 3.2A. It just gets hotter quicker..
 
You must log in or register to reply here.

Similar threads

dan1million
"Zero D" Mag 3xP7 Bike light using a HipFlex Driver...
Replies
17
Views
10K
javiole
J

Latest posts

Back
Top