Makita 18V Worklight Dead - DML185

[Thebigskysky]

Hi all, thanks in advance to any advice...

I have a Makita 18V work light that just kicked that can this evening.  I have been using it for a little over a year now and I have been pretty happy with it since it came bundled with the tools I actually wanted.

So things i have done, found out, tested, etc.:

• Tried multiple batteries, bulbs (both were still good), percussive maintenance
• Tested the switch, works fine, even bridged it to eliminate it as suspect
• Nothing on the board looks crusty, burned, obvious
• I have output voltage at battery voltage but when I apply it to the bulb, it drops to zero (no light)
• Once voltage is dropped from trying to apply it to the bulb, I have to either turn off than on the circuit via the switch or unplug the battery and re-plug it in if i have the switch bridged.

Components:
Component with flat facing the board edge - https://www.mouser.com/datasheet/2/308/on%20semiconductor_tl431-d-1195783.pdf
Transistor with flat facing the cap - http://www.unisonic.com.tw/datasheet/9015.pdf
Large transistor in the center of the board - https://www.mouser.com/datasheet/2/308/KSB772-1301146.pdf

Please let me know if you need more information, I cant figure out why the voltage drops as soon as i apply it to the bulb.  I would like to try and fix this if i can since im cheap.  Thank you!!!

[Thebigskysky]

More Pictures

[Thebigskysky]

More Pictures

[Armadillo]

Use a good battery for the test. Jumper any switch for the moment;

Use your hand to feel the components, any one of them is HOT to the touch?

Show a picture of the underside copper trace side of the PCB.

[fzabkar]

The instruction manual states that the xenon bulb is rated at 18V 0.6A:

https://cdn.makitatools.com/apps/cms/doc/prod/DML/2f0131ce-62c5-473d-930a-1fb58b2faf2b_DML185_IM.pdf

The product page states that the flashlight incorporates an over-discharge protection circuit:

https://www.makita.co.nz/products/model/DML185

I suspect that the PCB senses the battery voltage via the TL431 precision reference IC and then controls the lamp via power transistor Q2. I would think that the battery pack would have its own internal over-current protection device, eg a polyswitch.

[BurningTantalum]

I thought that these Makita packs actually flagged a dead cell etc and would not then charge on the Makita charger regardless of what one tried, apart from charging them externally, and the flag was permanent, so is it actually charging on the charger?
It would possibly be the pack that is dead, or at least one cell. It is very common. I would charge the pack, check the voltage, and try connecting the bulb direct to the terminals (as a load test.) I suspect that it will show the same effect.
If so, it is possible to take the top off these packs and determine which of the cells is at 0.0V. It is possible to graft in a replacement from another scrap pack to get it all running again. Just cut the dead one out with a Dremel and leave plenty of tag strip to solder to.
BT

[Thebigskysky]

In regards to the battery packs, i have two of them and both work with my other tools which even just freespinning must induce far more load on the battery than just lighting a lamp. 

I would test the circuit to see if any components get hot but as far as i can see, when ever i try to complete the circuit via the bulb, the voltage drops to zero so i dont think i could get it to even warm up if that was the case. 

I am guessing it has something to do with one of the three main components i listed, i just am not sure what would basically switch the power off when i connect it to the bulb.

Use a good battery for the test. Jumper any switch for the moment;

As for the switch, there is only one physical switch and i did the testing of the circuit with/without the switch bridged so im fairly sure it is not the culprit.

[Armadillo]

Hi;

When you said, the battery voltage drops to zero, do you mean the BT+ and BT- or the BU+ and BU- terminal? Did you measured by a voltmeter?

If it is BT goes to zero, then experiment to confirm Battery is no good, see attached. The current can be read. If zero or very small current, the battery is no good. [if big current, then your load is shorted].

If it is BU terminal then likely the TL431 is defective. To confirm, measure the voltage, Cathode of TL431 to BT- will be about same as BT+ voltage meaning it didn't pull the base of the Transistor Down enough. [or the big transistor itself is defective].

 

[MarkF]

I found a circuit diagram from here
   https://www.circuitsonline.net/forum/view/121168

Redrawn.  You need to verify that it is correct.

   

[Thebigskysky]

So yes, I tested voltage via the BU+/- and when they are loaded, they cut to zero.
At the same time i maintain voltage on the BT+/- lines and throughout the board.

When i load the circuit and it drops to zero, the base line on the B772, goes to zero but almost immediately recovers when i place the probe to check the voltage, which opens the gate and re-energizes the circuit.  If i do not probe the base, then the circuit will not re-energize as checked at the BU+ line.

So if the base line of the B772, is being driven to zero, which in turn would close the gate, my guess is the B772 is not the problem.

If that is the case, what would i be looking at for testing to narrow it down between the other two components? If my rational is indeed correct?

[Armadillo]

As said, if you load it and the base measure ~= to BT+

then the TL431 is defective,

The voltage you measured unloaded is phantom voltage, you just ignore it.

Load the circuit, then measure the base voltage.

if the base measure much lower than BT+ > 0.7V, then the big transistor is faulty or the lamp is open circuit.

[fzabkar]

ISTM that the "C" in "C9015" is an error. If the transistor is indeed a PNP type, then one would expect an "A" or "B", if anything.

That said, ISTM that the circuit works as follows.

Initially the TL431 is switched off, which means that Q2 is also off. When S1 is closed, battery current flows through D1, C1 and R2. The voltage across C1 is initially 0, and then rises as the capacitor charges. At switch-on, the voltage at the Ref pin of U1 is ...

Vref = (Vbat - Vfd1) x (R4 || R3) / (R2 + R4 || R3)

That's about 12V. This voltage turns on U1 which in turn enables Q2. R4 and R3 now form a potential divider which takes control of U1. In the meantime, C1 becomes fully charged and no more current flows through R2. If the battery voltage now drops below the over-discharge threshold, U1 switches off and Q2 is then latched off. Since C1 is fully charged, it must be discharged to reset the latch. This happens automatically when S1 is opened, ie base current flows from Q1 to ground via R1, thereby switching on Q1 and discharging C1.

The over-discharge threshold voltage is given by ...

Vdis = 2.5V x (1 + R4/R3) = 2.5 x 4.9 = 12.25V

[Zucca]

Toss it, or Ebay it as not working.

What you want/need in 2018 is a LED (better COB) portable light:

https://www.amazon.com/Rechargeable-Flashlight-1000Lumen-Floodlight-Construction/dp/B07DS34QN2/ref=sr_1_4?ie=UTF8&qid=1548766293&sr=8-4&keywords=cob+led+portable+work+light

if you want to use the 18V makita battery pack a simple and cheap DC/DC converter will do the job nicely.

[Thebigskysky]

Ok ALL.... sorry for the delay, apparently my last reply did not post where i said i was going to order some components and then head out of town for the weekend and get back at it on Monday...

Well, Monday came, so did my parts, ordered a couple of the TL431's and B772's, and got to work. 

I was just going to replace the TL431 and see if that fixed my problem but i got greed with a hot iron and decided to hedge my bets and did both.  Well, it works.   

Though i do understand this doesnt really tell me what was wrong, it doesnt matter, it works... but my guess is it was the TL431 was bad. 

Thank you guys SO MUCH for the quick responses and helpful tips and info... I look forward to growing in this community as i try to get further into circuit board / mobo repair as a hobby.  CHEERS!