Battery powered lasers - matching current

Discussion in 'Power Electronics' started by tothemoonn, May 21, 2018.

  1. tothemoonn

    Thread Starter New Member

    May 21, 2018
    12
    1
    Hi all -

    This is my first post so I will do my best to not break etiquette!

    I'm working on a simple electronics project that requires me to power two 3 volt lasers with a total of 2 AA batteries.

    The lasers are a bit of a black box. The company would only tell me they operate on 3 volts (they sell a no-frills, two AA battery pack to operate a single laser) and that they have some form of linear regulator inside. Combined they draw about 430 mA with a fresh battery, dropping down to a combined 250 or so mA at the point the batteries will no longer power the pair...and that's all the detail I have on the laser circuits. I've attempted to disassemble but they seem to be epoxied in place...no visible fasteners, no rotation for a threaded insert etc.


    One of the requirements though is that both lasers need to output roughly the same intensity light up until shutoff from a low battery. Currently one laser will begin to dim before the other when the batteries are about to fail. So Im wondering if there's some low-overhead method (keeping in mind that I cannot gain access to the laser circuitry, just the power leads) that I could use to ensure each laser receives the same current until the battery can no longer supply it and both lasers turn off?

    The person who developed the project ahead of me combined a BA30BC0 LDO and another part in a SOT-223-like package marked "TK 4DG" which I have not been able to identify, to address the problem. Im told that the lasers did in fact start failing to power basically in unison (at the cost of overall battery life) but Im too much of a novice to have a guess at what that part is/does. The addition of the BA30BC0 also strikes me as pointless given the internal LDOs....but again Im a novice with electronics and havent done anything battery powered. Aside from some obvious decoupling these are the only two parts outside the laser housings...there seem to be too many things that come in a package like that and Im not getting anything useful searching for TK 4DG. I have found mention of something called a current mirror, but I think it's not quite what Im looking for, and current mirror ICs I've found are for outputs on the order of 50 - 100mA.

    Thank you for an guidance!
     
  2. danadak

    Well-Known Member

    Mar 10, 2018
    1,758
    362
    For sure the preferred method of driving a laser is T compensated constant current. This will insure you do not shorten lifetime of them. Also contant current insures, to a first order equal flux output.

    Without access to internals its a bit of a guess as to what can be done here, and of course meeting specs on lasers impossible.

    Regards, Dana
     
  3. dl324

    AAC Fanatic!

    Mar 30, 2015
    7,245
    1,683
    Welcome to AAC!
    The reason one laser dims before the other is that their forward voltages are different. The one with the higher forward voltage will begin to dim sooner than the one with the lower forward voltage.

    The solution is to drive each laser at the same current. There's no low overhead way to do that. Operating from 3V complicates things. A current source needs some headroom and you don't have much with 3V.
     
  4. Hymie

    Active Member

    Mar 30, 2018
    532
    125
    1. Measure the input voltage at which the difference between the laser outputs become unacceptable in your application.

    2. Build a circuit that switches off the battery supply once the battery voltage reaches this level.

    (Note that once the circuit switches off, it should not switch on as a result of the battery voltage rise due to the drop in current draw)
     
  5. whitehaired novice

    Member

    Jul 15, 2017
    232
    28
    Would a "work around" of switching which set of power leads goes to which LED every so often help?
     
  6. tothemoonn

    Thread Starter New Member

    May 21, 2018
    12
    1
    Hi all - thank you for the ideas. I suspected that my main problem addressing this would be the fact that Im trying to get 3v @ like 400mA from a source that drops below 3V quickly. I'm going to investigate Hymie's suggestion and look into simple circuits to cutoff power below 2.4 volts (I have a TI development board for a switching regulator that has a low battery feature - Ill investigate that a bit and see if it's configurable.

    So the forward voltage differs just by virtue of manufacturing and that is causing the issue? Two laser diodes of the same brand in the same product will vary to a large enough degree to cause this?

    The lasers are in parallel so there's nothing really to switch unfortunately.
     
  7. dl324

    AAC Fanatic!

    Mar 30, 2015
    7,245
    1,683
    Yes. Below is the IV graph for a regular LED; it's essentially the same for lasers which, after all, are LEDs.
    EDIT: Referring only to solid state laser LEDs.
    upload_2018-5-22_7-47-26.png
    When the voltage drops to region where the slope isn't as steep, small differences in forward voltage result in larger drops in current. LED/laser brightness is proportional to current.
    Yes. Manufacturing variations give a range of forward voltages and brightness. Manufacturers sort (bin) them by brightness and give a range of forward voltages. To get lasers with matching forward voltages, you could ask the manufacturer to test them for you (which can be expensive and is only done for high volumes), or you could do it yourself.
     
    Last edited: May 22, 2018
  8. tothemoonn

    Thread Starter New Member

    May 21, 2018
    12
    1
    Great explanation, thank you! Sounds like it will be easier to just devise a circuit that will cut off both lasers at a voltage on the threshold of where this kicks in. The company making these lasers seems unlikely to do matching for us, and we'd still like to avoid disassembling the lasers to do it ourselves ( I also probably dont have precise enough instrumentation here ).
     
  9. Hymie

    Active Member

    Mar 30, 2018
    532
    125
    You could try something like this in the front-end supply to the lasers.

    https://www.ebay.co.uk/itm/3-3V-5V-...hash=item3f859fda21:m:mDY-vSHscoLlLJuNrlR6IPA

    The circuit claims to output 3.3V with an input voltage range of 0.8V - 3.3V; although there will be loses within the circuit, it may actually allow you to get a greater amount of energy out of a battery with it operating down to 0.8V. The circuit is likely to abruptly switch off once the battery is close to being exhausted (<0.8V).

    The ebay seller has not disclosed the output current rating for the device – so you might need to seek a higher power version.
     
    Last edited: May 22, 2018
  10. MisterBill2

    Well-Known Member

    Jan 23, 2018
    1,558
    276
    I have used laser packages from "Creative Technologies" (laser66) quite a few times, and they do contain a proprietary driver circuit that works very well. Of course my project powered them from a separate -5 volt power supply in a dip-style package. If you can change battery types, in the AA size package you could get 2 lithium cells and have plenty of voltage. That could be another option. But if you are locked in on the battry size and type, sorry about that.
     
  11. KeepItSimpleStupid

    Distinguished Member

    Mar 4, 2014
    2,879
    506
    Comments:

    1. Current draw may be too high for AA cells: https://en.wikipedia.org/wiki/AA_battery

    2. The Kentli batteries mentioned above might help

    3. Vf is likely different for each diode. This basically determines when it dies.

    4. At the same voltage, it's likely you don;t get the same intensity. Again, based on Vf

    5. You need to figure out what's going on. Measure V, I and intensity.

    Ideas knowing very little.

    Build a low cutoff for each laser based on say the LT6700 micro-power comparators so at least they go off at the same time.

    If you had access to the diode separately, a circuit that dropped a small voltage might equalize them.

    It MIGHT be possible to turn each laser on/off via PWM. The PWM duty cycle would equalize the intensity.
    e.g. 97% on for one Laser, 100% on for the other.
     
  12. MisterBill2

    Well-Known Member

    Jan 23, 2018
    1,558
    276
    One additional choice would be to get laser diode assemblies without the driver electronics included and set the current in the driver yourself. I did that on one prototype run and while it can be done it is a miserable task. So I don't recommend it.
    If you can tell us more about the application perhaps somebody can provide a better idea. That does happen here many times.
     
  13. tothemoonn

    Thread Starter New Member

    May 21, 2018
    12
    1
    Thanks everyone for the helpful suggestions - I definitely have some plotting to do...

    I may give one a try just in case but Im skeptical that I'd squeeze much more out since the current demand is pretty high.

    Unfortunately I was brought in on this project after the lab had already commissioned a custom housing be machined for these lasers in particular so I really am stuck with AA and these black box lasers.

    I think the LT6700 might be the best thing at the moment. I am locked into these lasers, and they wont have much control over battery brand in the field (Ethiopia) so it's seeming like all there is left is adding something like this to cut them out when the batteries drop to ~2.4V.

    As I mentioned above the lab had already spent a considerable amount having an enclosure made for these particular lasers so I dont think they'll be open to changing them or the AA battery thing. The device is used for determining the size of wild apes - it screws onto a DSLR and each laser is carefully set to parallel in the box so that the distance between the two points projected on an ape can be used for scale/sizing up the ape later. So matching the intensities doesn't need to be precise, but they'd rather it stop functioning at the point when the two points become drastically different.
     
  14. KeepItSimpleStupid

    Distinguished Member

    Mar 4, 2014
    2,879
    506
    There is an LT67xx- variant that's a single comparitor.
     
    tothemoonn likes this.
Loading...