Help building D-Star- Mtbr.com
Results 1 to 41 of 41
  1. #1
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28

    Help building D-Star

    Hey it is starting to get dark before 8PM now in Florida and is cutting in to my Wed night rides so now it is time to put together the beautiful housing Steve sent me this summer


    I will have my father to help me in mid October (former electrician/electronics engineer) because I am completely retarted about electricity (but I can solder very well!!-- I know it makes no sense). Thanks for helping a n00b out


    What is the most simple way to build is what I could use advice on. Cost is not really an issue between drivers & batteries

    1. order 3xMCE on 20mm stars in parallel or in series?
    2. which driver (maxflex looks sweet)
    3. and what type of battery pack (it depends on if the LEDs are in parallel or series right?)
    4. obviously Ledil CMC SS is the way to go for optics?

    I just need to get everything together by mid October.

    OFF TOPIC

    Just finished my first XC race today in the Florida State Championship Series. The trail was amazing and I finished 8th in my category (base 18 - 29 male). But I need to be able to ride at night to keep training!

  2. #2
    mtbr member
    Reputation: troutie-mtb's Avatar
    Join Date
    Sep 2007
    Posts
    3,062
    Hi Panabob

    if you are using the maxflex then you want them on stars individually wired as you need to be able to have 2 strings of 6 emmitters in parallel

    Maxflex is probably your best choice .

    a 14.8 volt li ion battery is also a good choice certainly no less .

    I would be tempted to go with 2 RS and 1 SS optic for a longer throw .

  3. #3
    mtbr member
    Reputation:
    Join Date
    Apr 2006
    Posts
    596
    Hi PB A 16.8 volts battery; li-ion is a good choice with the three MCE hooked in 2p6s. I have the exact set up and I (heart) it. A boost controller (like the Maxflex) is not kind to the battery, as the battery discharges, the current draw increases. So at 16.8 volts (battery fully charged) the light draws less than 1.5 amps, that increases to over 2 amps at 12 volts. Battery is discharged at 11.5 volts. My battery consists of eight 2400mAH 18650 cells in a 2p4s configuration. This is [email protected] battery, figure current draw of 1.75 amps average, over 3 hours run time on high. I seldom use it on high. it's a wall of light. I use 3 SS optics, now I'm gonna try 1 RS optic in the mix. It KILLS the HID lights. good luck

  4. #4
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    awesome thanks for the feedback Sir Troutie and MTBfreak, just what I needed to get started. I will post back in month if I need any further help. I have a 300 lumen torch I'm using for the meantime, it will be a crazy to then upgrade to 3xMCE lol

  5. #5
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    well my Dad flies in Friday, can I get some pre-guidance on how to wire the MCEs individually?


  6. #6
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    15
    Any help yet on the wiring front yet PanaBob,maybe Troutie or Dstar can help.

  7. #7
    mtbr member
    Reputation:
    Join Date
    May 2005
    Posts
    3,830
    Pretty tough to give wiring instructions without knowing what battery and driver he is using.

  8. #8
    mtbr member
    Reputation:
    Join Date
    Nov 2008
    Posts
    125
    As I understand, you're looking at a 4S 14.8v (16.8v peak) Li-ion battery, as an 18.5v 5S at full charge is about 21v, which is more than the 20v maximum for the Maxflex board. Using a smaller voltage battery with a Maxflex eg. 3S 11.1v will cause a greater current demand on the battery pack, be less efficient and may require additional heatsinking for the driver. So don't do it!

    DON'T TAKE ANYTHING BELOW HERE AS GOSPEL

    but as I understand for the LED's, you can either choose individually wired as per your drawing or have two series wired MC-E's and one individually wired, say in the centre. This cuts down the amount of spaghetti in the case. As long as you end up with 6 individual dies on each circuit 6S2P.

    So, for ONE circuit,

    (1) take one +ve from the Maxflex to +ve on the series star.
    (2) from the -ve from series star to the first +ve on the individually addressable star, lets say top right as per your drawing
    (3) Then a link from the corresponding -ve top left to the next +ve, 2nd right from the top
    (4) Now wire from -ve 2nd left from top back to the -ve on the driver.

    Repeat for the 2nd circuit.

    To wire individually addressable stars will increase the amount of linking cables but is perfectly possible. So working your way down one star:

    (A) +ve from driver to top right
    (B) -ve top left to +ve 2nd right
    (C) -ve 2nd left to +ve 3rd right
    (D) -ve 3rd left to +ve bottom right
    (E) -ve bottom left goes to the +ve on the next star, say top right again
    (F) -ve top left to the next +ve, 2nd right from the top
    (G) Now wire from -ve 2nd left back to the -ve on the driver.

    Repeat for the 2nd circuit.

    Hope that helps. With any luck the wise ones might see this and scream out if anything is wrong.

  9. #9
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    15
    Thanks AlisterG that sounds reasonable even to me ,do you
    know what sort of run time a 4000mah 14.8v lipo might give on the full setting
    Thanks in advance
    solarbake

  10. #10
    mtbr member
    Reputation:
    Join Date
    Nov 2008
    Posts
    125
    With all individual addressed stars - another way is to imagine dividing the three stars in half, horizontally. So there are 3 groups of 2 dies above this imaginary line and three groups of 2 dies below it. These will be your two separate circuits.

    Starting from the right star, working left, for the upper circuit above this imaginary line:

    (a) +ve from driver to +ve top right
    (b) -ve top left to +ve 2nd right
    (c) -ve 2nd left to +ve top right of MIDDLE star
    (d) -ve top left mid star to +ve 2nd right
    (e) -ve 2nd left to +ve top right of LEFT star
    (f) -ve top left to +ve 2nd right
    (g) -ve 2nd left returns to -ve of driver

    Then repeat for dies below the imaginary line. Six dies each on two circuits. This should generate more conveniently located wiring within the case.

  11. #11
    mtbr member
    Reputation:
    Join Date
    Nov 2008
    Posts
    125
    Using a maxflex at 1200ma, approximate runtime = (14.8v x 4ah)/(0.6a x 2 x 13v) = 3.8hrs.

    I have assumed that the MC-E Vf is 13v for series wired, some folk have got lower (better) Vf with recent MC-E's. So after other electrical losses, you should get 3hrs +.

    With M-bin MCE's thats kicking out over real 1700 lumens, allowing for 10% losses though the optics, providing you keep it cool. Expect over 1500 - that's bright. You'll find the 700ma setting will still push over 1000lm.

  12. #12
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    hey thanks for the help I am getting ready to put it together right now

  13. #13
    mtbr member
    Reputation:
    Join Date
    Nov 2008
    Posts
    125
    Correction to runtime: (14.8v x 4ah)/(1.2a x 6 x 3.5v) = 2.35, say 2 hours in real life.

    (slopes off to do 100 lines - I must check my calculations)

  14. #14
    mtbr member
    Reputation:
    Join Date
    Apr 2006
    Posts
    596
    Keep in mind, battery time is calculated at room temperature. As it get colder, the battery gets wimpy. Give yourself 15-25% LESS time at cold temps. Keep the battery next to your body to keep some heat into it.

  15. #15
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    15
    You sure that run times correct AlisterG as i thought the maxflex draws 1200 ma max
    so should give over 3 hrs on full.

  16. #16
    mtbr member
    Reputation:
    Join Date
    Nov 2008
    Posts
    125
    I'm sure, as this calc works out power. My first calc was incorrect as I used the wrong Vf. The Maxflex produces 1.2A for the LED's but draws between 1.5A to over 2A from the battery. The increase in current is due to the Maxflex demanding more amps to maintain the Vf to the LED's as the battery voltage decreases.

    This is the main drawback with boost circuits and the only reservation I have about this light. Were it able to house a Taskled Hipflex driver, (a buck circuit), it would be just about perfect for multi die LED's.

    Have a look at this, similar to LED Pro: http://www.kenworster.com/ledcalc/

    I put in Vf = 21v (6 x 3.5v), Vbatt 14.8v, I = 1200mA, driver efficiency 90%. Runtime came out at 2hrs 6mins, less than my rough calc.

    Don't forget, at full power this light will be drawing around 25 watts, with light output to match a 55-60w car headlight.

  17. #17
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    success. a massive wall of light just lit up my backyard!!!

    will post DIY pics in a day or two to help the next noob put this together

  18. #18
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    15
    Hi PanaBob what battery and run time you getting.

  19. #19
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    I had to build a battery bank out of eight 1.5v duracel batteries to test it last night. I just got my 14.8v 4400 mAh battery pack today so I will not be able to test run times till this weekend when I have time (I work a lot) and figure out how to program the maxflex

    I'll post pics to show how its going. Right now I have the 3 MCEs wired to the maxflex. I still have to heatsink the LCDs and maxflex, get the optics to fit in the casing, caulk/water seal it, and connect the battery

  20. #20
    mtbr member
    Reputation: HEY HEY ITS HENDO's Avatar
    Join Date
    Feb 2008
    Posts
    716
    Hi PanaBob, somehow missed this thread! d`oh
    got any pics of the build so far?
    i set the optics and stars before soldering
    i think you may have problems if your all wired up before you position/affix the mce!
    (click on pics for higher res.)



    ...Scun.thorpe, UK

  21. #21
    mtbr member
    Reputation:
    Join Date
    Nov 2008
    Posts
    125
    Much neater than my suggestions.

  22. #22
    mtbr member
    Reputation: HEY HEY ITS HENDO's Avatar
    Join Date
    Feb 2008
    Posts
    716
    Hey Alister i have an advantage, i`ve got one
    in the drawing, the long wire goes thro to the back ......

    ...Scun.thorpe, UK

  23. #23
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    15
    Hi PanaBob,
    hows that light going ,any more pics,info yet.

  24. #24
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28


    looks like I may need to shorten my wires or else tuck them really well so they do not touch the maxflex. You can see where the far left wire got burnt when I had the light on for 10 seconds. Also the far right wire seem to have started to smoke it is slightly discolored to white. More troubling is the R5 and R6 IC seem to have melted in the center with what looks like black tar?

    Did I wire this correctly? The light is bright as hell

    The wire on the far right of the switch is to the negative on the battery
    Last edited by PanaBob; 10-25-2009 at 07:08 PM.

  25. #25
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    Here is how we soldered the MCEs in parallel



    and the battery I am using for power

    https://www.all-battery.com/li-ion18...connector.aspx

  26. #26
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    Maxflex 4 reference
    Attached Images Attached Images  

  27. #27
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    well I think I figured it out, the battery input (discharge) current is probably too high at 6A when the max input current for the driver is 3A...

  28. #28
    mtbr member
    Reputation:
    Join Date
    Apr 2008
    Posts
    24
    R5 and R6 are just two SMT resistors and can be easily sourced and carefully replaced. Removing the old cooked ones will be the hardest part. Hopefully you haven't cooked the main transistor on the Maxflex driver circuit.

  29. #29
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    15
    H i PananBob
    Looks like the maxflex frazled to me ,it should have been wired as Alistair
    said lokk at a few posts before.


    With all individual addressed stars - another way is to imagine dividing the three stars in half, horizontally. So there are 3 groups of 2 dies above this imaginary line and three groups of 2 dies below it. These will be your two separate circuits.

    Starting from the right star, working left, for the upper circuit above this imaginary line:

    (a) +ve from driver to +ve top right
    (b) -ve top left to +ve 2nd right
    (c) -ve 2nd left to +ve top right of MIDDLE star
    (d) -ve top left mid star to +ve 2nd right
    (e) -ve 2nd left to +ve top right of LEFT star
    (f) -ve top left to +ve 2nd right
    (g) -ve 2nd left returns to -ve of driver

    Then repeat for dies below the imaginary line. Six dies each on two circuits. This should generate more conveniently located wiring within the case.

  30. #30
    mtbr member
    Reputation:
    Join Date
    Nov 2008
    Posts
    125
    I make no claim to any electrical or electronic knowledge, but Hendo's wiring should work, it can be considered as a series of six sets of two parallel resistors.

    What worries me on your photo of the LED's is the blobs of solder that appear to be touching the heat sink plate and that you have used bare wire, potenially causing shorting.

    As I understand the Maxflex will ground via the heatsink pad to the casing. Under normal circumstances the Maxflex will only draw the current it requires. Any shorting could cause it to draw much more current, frying the board.

    Again, I'd await the feedback from others to confirm. Hendo, this is a call to you!

  31. #31
    mtbr member
    Reputation:
    Join Date
    Apr 2006
    Posts
    596
    Hi there PB, I suppose the bad news first. The inductor got (square bit) so hot it unsoldered itself from the pcb, also the cooked resistors, also the cooked wire. These are bad signs! With all the MCE's soldered in parallel, their required current would be 8.4 amps. The Maxflex will limit itself to a max of 1.2 amps. This is not a problem (except the MCE's wouldn't be very bright). Since your MCE's are bright, you have killed the Maxflex. The gold heat pad gets glued to the heatsink, I hope the other wires didn't also touch the HS. The proper hook-up for the MCE's is 6s2p. The picture I enclosed will help you. Sorry about your loss.
    Help building D-Star-mce-6s2p.jpg

  32. #32
    mtbr member
    Reputation:
    Join Date
    Nov 2008
    Posts
    125
    Yep, that looks like each MC-E board is parallel wired to me, with each individual die in parallel too. 12P1S, no wonder it blew the t*ts off the board.

  33. #33
    mtbr member
    Reputation:
    Join Date
    Apr 2006
    Posts
    596
    In the electronics trade, we call that letting the smoke out! You have to understand that every piece of electronic gear has a small amount of "magic smoke" inside it. The gear will/may operate until you release the smoke. After that action, it's tango uniform.

  34. #34
    mtbr member
    Reputation: HEY HEY ITS HENDO's Avatar
    Join Date
    Feb 2008
    Posts
    716
    PanaBob, i see the only way to rectify this problem is to start over with another maxflex driver,
    thinner and shorter wiring, and unsolder and rewire the mce`s for maximum brightness,
    ...how are the mce`s fixed to the heatsink?
    ...Scun.thorpe, UK

  35. #35
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    15

    New question here.

    I Thought your meant to have two groups of 6 leds so you need two positive and two negative wires back to the maxflex.the above sketch is
    the above sketch right???????????
    Would of been nice to have some input from Mr Deestar himself about all this after all
    it is his light
    Last edited by solarbake; 10-26-2009 at 03:52 PM.

  36. #36
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    The MCEs are attached to the heatsink with AS5 which is pretty sticky enough to work with for now and they stay in place till I finalize them, the can epoxy a small part of each to the HS incase I need to remove them in the future.

    The LEDs should be fine, I will replace the maxflex and start over. At least I think I know what to do now

    thanks

    Also, there is no shorting from any of the soldering but I will go with higher gague wire, 24 I guess? I was using 22

    The gold heat pad gets glued to the heatsink, I hope the other wires didn't also touch the HS.
    no shorts I was careful there. The VIN + wire came out I think because it got hot. l found that the heavy annodizing does not conduct either, neither does the arctic silver 5. Obviously the untreated HS would conduct but I was careful there and it is not too close
    Last edited by PanaBob; 10-26-2009 at 06:01 PM.

  37. #37
    mtbr member
    Reputation:
    Join Date
    Apr 2006
    Posts
    596
    Hi SB, 6s2p that is the hook-up in the drawing, 6 in series and 2 in parallel. Only one wire is required (1 for plus, 1 for minus). The parallel connection is made on the LED's. Just hold your individualy connected MCE star up and wire it.

  38. #38
    mtbr member
    Reputation:
    Join Date
    Apr 2006
    Posts
    596
    Hi there PB, if you are only using 2 MCE's, just ignore the last one, #3. then you'll be hooked up 4s2p.

  39. #39
    mtbr member
    Reputation: HEY HEY ITS HENDO's Avatar
    Join Date
    Feb 2008
    Posts
    716












    ...Scun.thorpe, UK

  40. #40
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    thanks Hendo new maxflex otw appreciate it and will post some pics when its done

  41. #41
    mtbr member
    Reputation:
    Join Date
    Jan 2009
    Posts
    28
    Thanks everyone and Steve for your awesome housing. I got my new maxflex in and wired like Hendos (very easy to do once you have a visual diagram). This time no magic smoke
    oh yeah and it is super bright

Members who have read this thread: 0

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

THE SITE

ABOUT MTBR

VISIT US AT

© Copyright 2020 VerticalScope Inc. All rights reserved.