Shedding some light on the RPL auto shut-off

[pseudonomen137]

Recently there's been a bit of confusion about the RPL's auto shut off and what triggers it, so I tried to do some investigation. At first it was said to be to protect the laser from overheating, then Jack mentioned it having more to do with the batteries. Finally, someone (I think it was Aseras) mentioned getting longer runtimes with slightly used batteries. Well, it looks like they are all partially correct. Here's the situation as I understand it from the broken English:

The RPL auto shut-off is ultimately to combat the laser overheating, but it doesn't do this by a temperature sensor. As a laser diode heats up, it becomes less efficient and so the RPL circuit makes it draw more power to sustain output. Eventually it starts trying to draw more current than it was spec'd to, and that's when the auto cutoff triggers to shut down the laser temporarily.

How the actual driver circuit works on the RPL puzzles me, but I did learn that a fully charged battery results in both higher input and output powers for the diode/laser. Therefore, using a fully charged battery, or a higher capacity battery will cause the diode current to already be close to auto off levels, whereas letting the battery drain a little before use may allow unlimited runtimes.

I hope that helps all you RPL users.



PS: If you are considering an RPL, send me a PM as I may have some useful info for you.


PPS: I haven't had time to work on my site recently (Will be called A1 lasers and used for selling the random stuff that passes through my laser collection), but eventually I will get it online and be able to sell additional tailcaps to you RPL owners who would like to have both (I bought my RPL 300 with the adjustable endcap, and my RPL Blue 17mW with the non-adjustable endcap. I found I much prefer the non adjustable one, but I only have one of those endcaps for two lasers)

 

[Aseras]

yep, mine runs better after letting the battery drain just a bit. Much longer run time and everytime I've measured I get more 20-50 more mw after bleeding off a bit of charge on mine. I usually just charge the bat throw it in the RPL with the armed led glowing and leave it overnight and it's good to go.

I'd like to get a non adj tailcap too. Many times I've found myself comtemplating going over to our machine shop and finding some stock aluminum and firing up the lathe. I don't understand why Jack says the tailcaps are fitted per laser. There's nothing special about them.

 

[pseudonomen137]

I'd like to get a non adj tailcap too. Many times I've found myself comtemplating going over to our machine shop and finding some stock aluminum and firing up the lathe. I don't understand why Jack says the tailcaps are fitted per laser. There's nothing special about them.

I think he means that they can normally only get one tailcap per laser. That's how it was explained to me when I tried getting an extra once. I aim to change that shortly though   (although it will be a bit on the pricey side as they have to be ordered specially from the company in China that originally makes the cases)

 

[me_too_lazy]

cant wait til your site is up ;D

 

[Aseras]

I've talked to Jack ( a few times ) and tried to get one too. What I don't understand is why he can't get them being a exclusive franchise/dealer, especially if you can...

 

[Deleted member 10578]

Questions, questions.
Here are the answers to what I know.
Perhaps the battery protection circuit was originally designed to prevent diode overheating, but in my opinion after running several experiments with the RPL and all of it's circuits "dead bugged" so to speak, that's not the case.

Here's my reasoning for this.
1. The 2.5 Watt n-Light IR diode in the RPL is rated at 2.5 Amps of sustained operating current ( I didn't see in the spec that it needed to be heatsinked either). When the RPL has a non-adjustable tail cap or if it has the adjustable tail cap and is on current level 9 it uses 2.0 amps of current; so it is only being ran at 80% of it's rated capacity. The heatsink in the RPL is not the entire head that you see on the outside. Look at Aseras's pictures, the diode is in that aluminum assembly and doesn't contact the outer shell directly; however it is in contact with this assembly which does give a fair amount of heatsinking capability.

2. I've tested current draw and the current doesn't increase overtime. In fact it decreases as the battery voltage decreases as it drains (ohm's law I=E/R shows us that).
3. Psuedo is correct, there is not heat sensor in the shutdown circuit. What there is is an Seiko S-8211C battery protection IC for rechargeable lithium-ion cells.
The S-8211C senses the battery voltage during use as well as monitors for large power fluctuations (ESD), when the battery voltage decreases to a sense level that is fixed inside the S-8211C (there are many variations/models to this part that allow triggering at higher or lower battery voltages) When the battery voltage gets low enough that it triggers the battery protection circuit. It shuts off the current by way of two N-channel Mosfets (very low on state impedance/resistance). It lower the gate voltage to the mosfet and prevent current from flowing.

What is happening with the battery is that under a load the battery voltage will slowly decrease as the cell is draining. The higher the load, the faster the voltage drops.
To get better / longer run times or lengthen the time until circuit shuts it down, you can do a couple of things.
1. Don't use protected 18650 cells, you don't need them with the RPL as there is already protection built into the laser itself. Using a protected cell add more impedance to the circuit and that reduces/limits the voltage/current to the diode.
2. Get the highest capacity cells you can find. This is the most important factor. The higher it's rated capacity, the longer it will run at a given current before reaching the battery cutoff voltage.
3. Make sure the cells are not too old. Rechargeable lithium-ion cells are good for between 300-500 charge/discharge cycles or 12-15 months. Even if they are rarely used, the cells age over time at after about a year their voltage will drop much faster than they did when they were new. Heat affect cell aging in a negative way.
4. Make sure the cells you get are of high quality, there are a lot of cheaper 18650 cells going around (in fact that is what caused those laptops to catch fire and be recalled last year). I find the LG cells to be the best I can find and they are highly tested and undergo rigorous testing at the LG factory.

I'm not sure why Aseras's cell works best when it is partially drained. I could see that if the diode current was overstaurating the crystal when fully charged, that output could increase as the battery drained down to a level where no IR saturation of the crystal were occurring. I would have to see and test his laser along with his battery to figure that one out; however I've ran hundreds of cells on as many RPLs and what he sees is not the norm.

One more thing, you could always bypass or disable the battery protection and get 100% run until the battery is too low to power the laser, but that could be dangerous if the voltage gets too low as it could cause the cell to explode.

I'll post more tonight as to about the tail caps and why I haven't been able to get them. Right now I need to take my son to the doctor.

I do know, that to get the most power from an RPL, get the adjustable as it allows you to prevent oversaturation of the crystal and much longer longer run times (usually continuous)... I guess that's why about 85% or more of the RPL sales are of the adjustable type.

Jack

 

[pseudonomen137]

I'm sorry Jack, but although I see the logic behind what you're saing, I have to completely disagree.

First off, some of the things you say assume that a 2.5W IR diode does not require any heatsinking. Just think about that for a second and tell me if that makes a shard of sense.

Also, I have not disassembled an RPL far enough to examine the driver circuit, but I find it highly likely there would be a buck convertor in there, which means that readings of the input amps can't tell you what's going to the diode.

I've felt the internals of the RPL 300 and my RPL blue when running at it does get pretty hot. The manufacurer has told me 30-50 seconds of runtime is recommended on the high power models, and that seems relatively accurate.

As I've mentioned on the phone, your theory doesn't quite make sense. You explain the shut-off as an undervoltage protection mechanism. That would mean that on the very first auto off, the battery would need to be replaced before the unit could be used again. That is not the case, and it wouldn't make sense for it to kick in after just a few minutes either.

Or at least that's how I see it. It would be interesting to hear your opinion Aseras.

 

[Rasel]

Wait sooo... the RPL will turn itself off if it starts to run to hot? :-? Never knew...

 

[Things]

yup, and when it starts drawing more currnet than usual

 

[Rasel]

Would you trust turning it on full and walking away?

 

[Things]

well it depends on the duty cycle it has...

 

[Rasel]

Well... throw out some examples

 

[Greenie]

Interesting. I never knew the RPL had an auto shutoff and now I want to know how it works.

 

[darklandz]

after browsin thru the opto homepage, i couldnt quiet make out what the non/adjustable tailcaps are/what theyr for/what they do?

 

[Rasel]

They change the amount of power emitted by controlling the voltage methinks

 

[Deleted member 10578]

I'm sorry Jack, but although I see the logic behind what you're saing, I have to completely disagree.

First off, some of the things you say assume that a 2.5W IR diode does not require any heatsinking. Just think about that for a second and tell me if that makes a shard of sense.

Also, I have not disassembled an RPL far enough to examine the driver circuit, but I find it highly likely there would be a buck convertor in there, which means that readings of the input amps can't tell you what's going to the diode.

I've felt the internals of the RPL 300 and my RPL blue when running at it does get pretty hot. The manufacurer has told me 30-50 seconds of runtime is recommended on the high power models, and that seems relatively accurate.

As I've mentioned on the phone, your theory doesn't quite make sense. You explain the shut-off as an undervoltage protection mechanism. That would mean that on the very first auto off, the battery would need to be replaced before the unit could be used again. That is not the case, and it wouldn't make sense for it to kick in after just a few minutes either.

Or at least that's how I see it. It would be interesting to hear your opinion Aseras.

Hi Pseudo,
No need to be sorry, that's what the forum is all about, the exchange of information and ideas.
A few months back I emailed you the spec sheet for battery protection circuit as well as the mosfet switches the battery protection circuit uses to shut off the current when the battery voltage drops too low.
If you no longer have the pdf of the spec, you can download it from this link and read more about how chip works to sense the battery voltage and cut off the current. These chips are mostly used in the lithium-ion chargers, but work for this application as well.
ftp://ftp.sii.co.jp/pub/ic/spd_dtst/dt_sht_e/lithium/S8211C_E.pdf

On the heatsink question, I concede you are probably right, I was typing faster than I was thinking as I was hurrying to get info typed in before my son's doctor appointment. I just looked closely at the diode spec and it didn't mention any requirement for a heatsink, but considering it's very small size, 2.5W is quite a bit of power for such a small device. I am sure they do get hot, but considering that they are only running at 80% when full diode current (level 9) is selected, I think they are doing just fine. From my tests, most of the adjustable RPLs run best with maximum visual output at current level 4 which is ~1.0Amps. At 1 Amp, the diode is only running at 40% of it's rated current.

Very interesting your buck converter idea.
I'll draw a schematic of the driver board and see if it looks like a buck converter, I didn't think of that. After I do the schematic capture, I'll put it into PSpice and see how it simulates.
Now that I think about it, when I was tinkering around with the battery protection circuit, I did measure current through the diode and it was the same as the current from the battery, at least within a few milliamps. These few milliamps are used for the battery protection chip I expect. So this means that basically diode current is ~equal to battery current. If you do measure current use a method that is non-invasive to the circuit. I soldered about a 1 inch length of 20 ga. wire between the base of the aluminum diode assembly and the driver circuit board and connected a LeCroy AP015 DC current probe around the 20 ga. wire. then plugged the AP015 current probe into my scope. To measure the battery current. I took this same size wire and soldered it to the positive terminal of the battery with silver solder and the other end to the base tail cap brass battery contact. To make the connection between the body of the tail cap to the body of the laser I used two large alligator clips soldered tail to tail and connecting one end of the two headed alligator clip to the laser body and the other to the tailcap body. Current was measured the same way by putting the current probe around the 20 ga wire connected the the battery.

I was told that the runtime should be 30-50 seconds right when they sent me the 1st prototypes. In fact, when I got the 1st RPL prototypes and I didn't know about tuning the diode current (I thought the more diode current the more output power, so I always set it to level 9) and at the time was only using 2400mAhr cells; two of prototypes shutdown after about 50 seconds or so. I emailed them and said something was wrong and they told me this was normal operation...
I told them, there was no way I could sell these long term with that short duty cycle and got them to agree to design changes to fix the runtime issue. They changed something as beginning around January of 2007, the run time increased dramatically.
If fact adjustable RPLs that are set on current level 4 or 5 will run at their highest output power and do it continuously..ie 30 minutes.

On your last question about the 1st auto off.
Try this experiment as I have.
Get two brand new cells and fully charge them.
Put one cell in the laser and run it until auto off happens(use diode current level 9 or a non-adjustable RPL so it will eventually shutdown). When auto off happens, shut the laser off and wait 5-10 seconds and turn it back on again and note how long it takes to auto off again. As soon as it auto offs, quickly swap the battery in the laser for brand new fully charged one and see how long it takes to auto off. This will show that it is mostly if not completely Dependant on the battery rather than the laser or diode heat.

The reason you can have auto off happen and then wait a few minutes and have it work again is that when the battery is under load it eventually hits the voltage level that causes the auto off. When the laser it turned off for a few minutes the battery recooperates itself somewhat. This too can be proven by making a test fixture that switches a 2 ohm (10 Watt) resistor across the battery and then measure the battery voltage under this load after 3 minutes. Then shut off the circuit and wait a few minutes or so and then switch the circuit on again and right away measure the voltage. What you measure will not be as low as it was at the end of the previous 3 minute run on the resistor.

These are fun experiments to run. Thing is I've been to busy to do a lot of this lately, there are lots of other things I want to measure as well, just not enough time.

I can't remember the last time I've had such a stimulating conversation in a thread, Thanks.

When is a good time to call you Carter?
How about tomorrow morning?
We could take some of this offline as I'm a slow typist.
I can also explain to you how to disassemble your RPL without damaging it (provided your work area is ESD protected) and more importantly how to get it back together again.

Jack