Page 1 of 2 1 2 LastLast
Results 1 to 25 of 31
  1. #1
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671

    4S1P battery pack problem

    OK, this pack has been a PITA since I bought it. Here's a thread to explain what I have:
    Wiring a battery pack to facilitate balanced charging

    Got everything done and charged up, hooked it to the light and...

    It works fine in low and flash mode but when clicked on high it flashes once and goes out.
    Swapped in a different pack and everything is hunky dory so it isn't the light.
    Voltage on the balanced leads are 4.09V, 8.20V, 12.32V, & 16.42V. Sounds OK.

    Any ideas?
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  2. #2
    mtbr member
    Reputation:
    Join Date
    May 2005
    Posts
    2,255
    Overcurrent protection on the PCB tripping. See post 13 in this thread.

    Troubleshooting light system build

  3. #3
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    Quote Originally Posted by Vancbiker View Post
    Overcurrent protection on the PCB tripping. See post 13 in this thread.

    Troubleshooting light system build
    Did that test; hooked it up & switched it to high-Flashed and went out.
    I checked voltage using the balanced plug and it was still at 16.42V.
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  4. #4
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    Quote Originally Posted by roxtar View Post
    Did that test; hooked it up & switched it to high-Flashed and went out.
    I checked voltage using the balanced plug and it was still at 16.42V.
    You can't check it using the balance plug, that's looking at the voltage of the cells, not what's coming out of the protection PCB. You have to check the voltage on the wires going to the light.

  5. #5
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    Don't really have a way to do that.
    Also, your recommended fix in that thread is way beyond my abilities.
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  6. #6
    mtbr member
    Reputation:
    Join Date
    Jul 2007
    Posts
    1,124
    Did this pack work properly before you went and added your balance leads as you posted in that other thread?

    I didn't see a pic in that thread of your balance lead all soldered in - you might post a pic of that here, just to make sure everything looks right.

    Can you go and remove your balance leads (and i mean unsolder them - should be easy to do and reattach down the road) and see if it acts any differently?

    If it still acts up - try bypassing the protection board completely. If I'm thinking properly - you should be able to solder the battery plug leads to the B+ and B- pads on the PCB, which will bypass the protection circuits for the time being. (the ones labeled Red and Last Black in Hendo's pic here)

  7. #7
    mtbr member
    Reputation:
    Join Date
    Jul 2007
    Posts
    1,124
    Hold on - quick question!!!!

    What driver are you running? I just remembered you were asking for help with wiring one of the 7135 linear drivers the other day. You aren't plugging this battery into that driver are you?

  8. #8
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    Quote Originally Posted by adrenalnjunky View Post
    Hold on - quick question!!!!

    What driver are you running? I just remembered you were asking for help with wiring one of the 7135 linear drivers the other day. You aren't plugging this battery into that driver are you?
    No, that's a different light. This is for the one in the "Very Easy XML Build" thread.

    Quote Originally Posted by adrenalnjunky View Post
    Did this pack work properly before you went and added your balance leads as you posted in that other thread?

    I didn't see a pic in that thread of your balance lead all soldered in - you might post a pic of that here, just to make sure everything looks right.

    Can you go and remove your balance leads (and i mean unsolder them - should be easy to do and reattach down the road) and see if it acts any differently?

    If it still acts up - try bypassing the protection board completely. If I'm thinking properly - you should be able to solder the battery plug leads to the B+ and B- pads on the PCB, which will bypass the protection circuits for the time being. (the ones labeled Red and Last Black in Hendo's pic here)
    I never tried it before adding the balanced leads and didn't take pics before re-heat shrinking the pack.
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  9. #9
    mtbr member
    Reputation:
    Join Date
    Jul 2007
    Posts
    1,124
    Unfortunately, I think you're going to need to open the pack back up and do some testing.

  10. #10
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    Quote Originally Posted by roxtar View Post
    Don't really have a way to do that.
    Also, your recommended fix in that thread is way beyond my abilities.
    You can measure the voltage on the P+ and P- terminals on the protection PCB.

    If you're talking about adding a capacitor to the output of the protection PCB, that's not going to help you in this case. That helps if the protection PCB current limit is being tripped by the in-rush current when the light is first plugged in. That's not what's happening here. You have the light running and then it cuts out when you go to high.

    The most likely thing happening here is the steady state current used by your light on high is simply higher than the limit set by this particular protection PCB. I looked back at the ebay listing where you bought the battery, and they just don't say what the current limit is for this pack. How much current does your light use on high? Can you give more details about the light?

    I can tell you how to change the current limit on that PCB, You can double it by simply adding a second R010 surface mount resistor in the open slot next to the existing one on the circuit board. But from what you said, you probably don't have a soldering iron. You can set different values by changing the resistor values, but you would have to check the current rating for some of the other parts on the board to make sure you're not going about the limit of the other components.

    There are also other things that could be tripping the protection PCB, but they are less likely so I'm not going to go into those details yet.

  11. #11
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    Actually, you're soldering on balance connectors, so you must have a soldering iron. If you can post all the tiny numbers written on each of the black chips on that protection PCB, I may be able to tell you the current limit for that board. Depends on if they are standard chips or have been customized.

  12. #12
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    Quote Originally Posted by MtbMacgyver View Post
    Can you give more details about the light?
    This lite:Very Easy XM-L Build
    using this driver:T6 3-Mode Regulated LED Driver Circuit Board for DIY Flashlight (2500~3000mA 4.5~18V) - Free Shipping - DealExtreme
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  13. #13
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    ok, that should be less than 1A average current draw from the battery and it would be really rare for the current limit to be less than 1A.

    Since that's a buck driver and you're dealing with a large voltage difference between the LED and the battery pack, it may be the pulsed nature of the current draw from the battery pack that is throwing off the protection PCB.

    It also could be some problem where the protection PCB thinks one of the cell voltages is dropping below the limit.

    Unfortunately, you're getting into stuff where I'd have to use an oscilloscope to figure out what's really happening.

  14. #14
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    When you say the light works fine with another battery pack, is the other battery pack also a 4S1P pack? I'm wondering if the driver is really happy with an input voltage that high. I know what DX says the specs are, but .......

    It would still be good to verify the voltage going into the light and know if it drops to zero which would indicate the protection PCB has tripped. One "trick" for measuring the voltage is to stick a couple of sharp sewing needles into the insulation of the red and black wires going to the light and then attach the meter leads to the needles. If you see 14.8 to 16v when the light is running on low and then it drops to zero when you switch to high and the light goes out, you've proved the protection PCB is shutting it down. If the voltage stays, then the driver is shutting down. Just be careful not to let the needles or meter leads touch each other.

  15. #15
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    Yea, the other pack is a 4S1P pack I made using protected cells and a simple, non PCB, 4 cell holder.
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  16. #16
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    OK, this is getting even more f*cked up.
    First off, I opened up the light so I could measure the voltage in when it was switched to high.
    Yes, it drops way down so apparently the protection circuit is shutting it down as we suspected.

    Just for grins, I also brought in another light I built, identically to this one; same LED, same driver, except with an in-line switch instead of the wired in switch. I also brought in the other 4S1P battery pack I built. This pack has no PCB board but is made with protected cells.
    In an attempt to make this easier I'll label the players. The new light/pack I brought in will be now known as good lite/good pack. The one I've been having the problems with are now bad lite/bad pack.

    Now the weird part;

    I plugged good pack into bad lite and everything worked fine. OK, that confirms it's bad pack at fault, right?
    However, I then plugged bad pack into good lite and it also work fine.

    OK, just throwing darts here; is it possible the wired in switch on bad lite could have something to do with this? Perhaps putting a small load that messes with the protection curcuit in bad pack, but not the protected cells in good pack?
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  17. #17
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    Oh well, just blew up that theory.
    I now hooked bad pack directly into bad lite, bypassing the switch.
    Still shut down on high.

    So, I have two identically built lights, both using the same LEDs and the same drivers.
    This pack works fine with one and not with the other.
    One light is bad, right?
    Except that, with a different battery pack, they both work fine.




    I hate electronics with the white hot hatred of a thousand suns.
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  18. #18
    mtbr member
    Reputation:
    Join Date
    Jul 2007
    Posts
    1,124
    meh - that's why I love this stuff. This kind of stuff can be done sitting in a comfy chair, indoors with a beer in hand.

    Replacing the CV joints and a-arm bushings in my truck last weekend, in an open carport, 38 degrees - that was the type of stuff I hate dealing with.

    If you have a multimeter, I'd test the current that is being delivered to the LED in each light - desolder one of the leads, and place a multimeter in current testing mode between the wire and the pad it was soldered to. get those readings with both packs on both lights.

    then, I'd do the same test for the current flow going into each driver in the same manner. that might help diagnose any differences.

    This is why I test everything on the bench before sealing up lights, bonding LED's to housings, taping drivers to heatsinks, etc.

  19. #19
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    You're likely dealing with a setup that's on the hairy edge of working or not working. If I were to guess, I would say the difference is the driver in the 2 lights. That's one of the problems with these really cheap electronic boards. You see a fair amount of variance from one board to the next. In this case you're dealing with cheap drivers and cheap protection PCBs. Introduce another variable into the mix, such as low temperature, and you very well may see a different behavior such as none of the combinations work. If you really wanted to prove this, you could swap the driver boards between the two lights.

    Given what you're seeing and have tested, I'm fairly confident that it is the over current protection circuit tripping on the protection PCB. Some brands of cheap protection PCBs do not deal well with the pulsed current draw that you get with switching LED drivers. They tend to react to the instantaneous current instead of the average current over a few milliseconds and as a result the noise transients set them off. The one driver that doesn't work with the one battery pack is likely generating more noise than the one that does work. The protection PCB is probably really close to tripping with both lights, but the one with more noise puts it over the edge.

    This is a case where adding an additional capacitor to the output of the protection PCB may take care of the problem after-all. That's because it'll smooth out the current flow. You would solder the cap directly to the P+,P- terminal on the protection PCB. A tantalum capacitor is best because they are very small and perform better than other types, but really any type will likely work. If you have a radioshack nearby, go pick up one of these and solder it to the protection PCB and see if it fixes the problem. Make sure the + lead on the cap goes to the P+ terminal, and the - lead on the cap goes to the P- terminal on the protection PCB. This type of cap will typically have a band on the side of the body denoting that lead is negative. It's better not to short the P+ and P- together in the process of soldering, but if you do then the protection PCB should trip and prevent anything bad from happening. You may have to charge it through the P+, P- terminals to reset the protection PCB if you short it. Charging through your balance connector will not do a reset.

    47µf 35V 20% Radial-lead Electrolytic Capacitor | RadioShack.com

    22 uF Electrolytic Capacitor : Electrolytic Capacitors | RadioShack.com

  20. #20
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671


    and




    Huge thanks, guys.
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  21. #21
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    I'm assuming that fixed the problem. If so, I'm not completely surprised it worked, but you connected the capacitor to the input of the protection PCB (B+, B-) instead of the output (P+, P-). The way you connected the cap will likely cut down the noise some, but it should be even more effective if the capacitor is on the output (P+,P-).
    Last edited by MtbMacgyver; 01-16-2013 at 06:54 AM.

  22. #22
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    Quote Originally Posted by MtbMacgyver View Post
    I assuming that fixed the problem. If so, I'm not completely surprised it worked, but you connected the capacitor to the input of the protection PCB (B+, B-) instead of the output (P+, P-). The way you connected the cap will likely cut down the noise some, but it should be even more effective if the capacitor is on the output (P+,P-).
    Sheeeit, your right. Wasn't paying attention.
    Here I was so proud of myself.

    Damn, I've spent as much time on heat shrinking this thing as the whole rest of the build.
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  23. #23
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    Quote Originally Posted by roxtar View Post
    Sheeeit, your right. Wasn't paying attention.
    Here I was so proud of myself.

    Damn, I've spent as much time on heat shrinking this thing as the whole rest of the build.
    Well, there's nothing wrong with leaving it that way if it works. If you've already re-wrapped the pack you may just want to see if you have any problems and move it if you do.

  24. #24
    mtbr member
    Reputation: roxtar's Avatar
    Join Date
    Sep 2009
    Posts
    671
    Quote Originally Posted by MtbMacgyver View Post
    Well, there's nothing wrong with leaving it that way if it works. If you've already re-wrapped the pack you may just want to see if you have any problems and move it if you do.
    No such luck.

    Remember this thing?
    Quote Originally Posted by MtbMacgyver View Post
    Introduce another variable into the mix, such as low temperature, and you very well may see a different behavior such as none of the combinations work.
    Bingo.

    Tried it last night in <30* and had to ride in low mode.
    Time to strip it down again and solder the cap to the correct leads.
    I'll be soldering to the main + & - outputs, right?
    Democracy is two wolves and a sheep voting on what's for dinner.
    Liberty is a well armed sheep, contesting the vote.

  25. #25
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    754
    Quote Originally Posted by roxtar View Post
    I'll be soldering to the main + & - outputs, right?
    Yes, the main outputs (P+, P-) right in the middle of the protection PCB. The same output that go to the main connector to the light. You can also use some short insulated wires to make the connection to the cap. That way you can tuck the cap in a more convenient spot such as next to the li-ion cells instead of on the top of the pack. Just keep them short and don't use tiny wires. Something the same size as your balance connector wires should be fine.

Page 1 of 2 1 2 LastLast

Posting Permissions

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