QIII + Dorcy 1AAA circuit + R123 (Li-ion)?

Is anyone running this combo? I just got a second QIII and finally got my hands on 2 Dorcy 1AAA LED lights (unfortunately both only have the less efficient AM50 chip, not the CX chip). I am planning on doing the same mod that Milkyspit did by swapping in a modified Dorcy circuit into the QIII. I also have some 750mAh R123 (they're more like 650mAh) Li-ions that I'd like to use with it. I haven't opened up the head of the new QIII yet, so I don't know the Vf bin, but if need be, I have some TX1K and TY0L stars that I could swap in.

Will the Li-ions fry the Dorcy circuit fresh off the charger (4.18v)? Or will it depend on the Vf of the star? I know the stock circuit with a SY0K star can handle the Li-ion (that's why I'm currently using), but I don't know if the Dorcy (stock or modified) can handle it.

I'm thinking about running this combo too, so anyone's comments/experiences are appreciated.

thanks.

I have a Q3 with the Dorcy circuit and a TWOJ. I have ordered a couple R123s and a charger, and they should be here in about a week. I'll try to remember to report my findings.

Hey Avro! /ubbthreads/images/graemlins/wink2.gif I've been pimping-out Q3s for myself and some other CPFers for a while now. Haven't used an R123 in any of mine yet, but I've installed plenty of Dorcy 1AAA circuits with no problem. On a regular 123 cell a TYAH star usually generates output that's as bright or brighter than stock, with 50-100% longer runtime than stock and WAY more graceful rundown at the end of battery discharge... I fell asleep one night while testing one, and woke up surprised to find that 8 hours into the elapsed runtime, the thing was STILL producing enough light to navigate easily through my house without tripping... that's no simple task given that I've got three sons, aged 5,3, and 6 months, and the floor is usually a minefield of toys! /ubbthreads/images/graemlins/ooo.gif

Anyway, that's probably not so relevant to you. I do know this, though: according to Doug S (at least I think it was him), the circuit should be capable of running with up to 4.5V input, so an R123 ought to do fine. I'd be more worried about the output side cooking the LuxIII... but my guess is that TY0L should have no problems at all!

Keep in mind, though, that the output on this circuit is Vf-sensitive, meaning the lower the Vf of the LuxIII, the higher the current flow will be. That makes the mod a little bit of an art... you kinda want to pair the circuit and power source with an emitter that drives as much current as possible through the output on a fresh cell... but not too much! /ubbthreads/images/graemlins/icon15.gif

Patience is DEFINITELY a virtue here. /ubbthreads/images/graemlins/grin.gif

Tell me about the AM50 vs. CX chips, please... that's new info to me! PM me if you don't want to pollute this thread. For my part, I'll pass along a couple tweaks that Doug S suggested for a little better efficiency at high output levels... first, add a third 10uF capacitor to the empty pads on the board... and second, short out the 1 ohm resistor sitting just above (as in closer to the LED) the inductor. These two tweaks should add several percentage points of efficiency when driving high currents out of the circuit.

Incidentally, for those who are new to the Dorcy 1AAA circuit, one of my favorite mods so far... the one-of-a-kind MilkMite (described in a thread elsewhere on CPF)... is powered by a Dorcy 1AAA circuit... here's a runtime graph...



Your mileage will vary based on 123 vs. R123, the Vf of the emitter, etc., but it's pretty safe to say you'll see MUCH improved runtime characteristics over the stock circuit!

Okay, I'm completely out of control at this point, rambling all over the place. Hope all this crazy info helps folks at least a little! /ubbthreads/images/graemlins/grin.gif

Milkyspit,

Thanks for the input.
As for polluting the post with info about the differences between the AM50 vs. CX chips or how to improve the efficiency of the dorcy board, I think that anyone who is checking out this post would probably be interested in seeing those topics discussed in more detail (I know I would.) Of course, I'm a newbie, so if I'm mistaken, and this is better handled in PM, maybe you and Avro could PM copies of your discussion to me as well.
Thanks,

--workin' to pimp my new Q3--

woolfam, keep us updated when you get a chance to test the Li-ions. And if it's convenient, could you check the chip that on your Dorcy circuit? It should be either AM50 or CX xx.

Milkyspit, it was your "My pimped out Q3" thread that got me started on this. /ubbthreads/images/graemlins/grin.gif And I was also reading Doug S' thread about the AAA circuit and the wonderful efficiency it posses for a cheap $6 circuit. However, after thoroughly reading his thread it seems that there are 2 different versions of the Dorcy circuit, an older one that use CX xx chips, and newer ones that used the AM50 chip, the difference being that the newer AM50 is less efficient, e.g. at Vin=1.5V, the efficiency dropped by about 18%. Unfortunately his test data only went up to Vin=1.5V, so I have no idea what the efficiency is like at Vin=3.0V or 4.2V (which is what I'm really interested in). That's why I was asking about what chip you had on your Dorcy circuit. If you got your results using the AM50 version, then that would be really good because even though it's not as efficient as the CX version, it's still almost twice as good (runtime) as the stock Q3 circuit. If your results are from the CX version then it's a different story because the results would not be the same. And the other sucky part is that there is no reliable way to determine which chip it contains without removing the circuit. Although there are 2 version of the light (one with diamond checkered rubber grips and one with long vertical rubber strips), the AM50 was found in both (I have 2 with diamond grips and both have AM50 chips).

Also, regarding the voltage input, I just found another thread by Doug S where he suspects that the Dorcy circuit can handle Vin & Vout of at least 6V, so I should be okay there using Li-ions. What would you guesstimate be for the best Vf bin to use with a 3.7-4.2V Li-ion? I know you're running H-bins, but that's with primary cells. I think I'll try the K-bin and see how it works out. It should run a tad longer because of the lower Vf, provided I don't run it too long continously. And I was going to do those tweaks to the circuit, but am having difficulty sourcing the 10uF capacitors. The dude at the electronics shop said they only had micro-puff(?) rated capacitors in the small 0805 package size (I think), the 10uF stuff was a small cylinder type, not the super small rectangular SMD component size. They had the 0805 sized 10 & 20 ohm resistors (for the 2-stage switch mod I'm going to do), but not the capacitors, weird. I'll have to check another shop.

Hikaru, no worries, the AM50/CX chip discussion is entire relavent to this thread. If the AM50 really sucks at the Vin range of 3-4Vs, then there's almost no point in using the Dorcy circuit over the stock circuit. I am primarily going to run my Q3s using the unprotected Li-ions because they were waaaay cheaper ($3 each vs. regular Duracell or Energizer 123s that are $7US each here in Canada) than primarys, and can be recharged (cheap to run). But I do want the flexibility to have decent run times if I do use primary cells (when travelling for instance).

My circuit has the AM50 chip. However, I have another (older) Dorcy at home that might have the CX chip. I'll check it out and let you know.

Looking for some more info on this combo. Thanks.

Hold the presses. In my case, ignorance is bliss. Milky modded my QIII with this setup and in my best scientific CPF jargon, this thing is cooler than snot on a block of ice.
My contribution was to get the lights to him by whatever means I chose.
And I did.
Vhyper007

I checked at home this weekend and I did have a Dorcy w/ the CX chip. (Actually, my son had one and I -traded- him for it) I put the CX version in my Q-III. My rechargeables should be here sometime this week.

I just received my rechargeables. They are reading 4.02v off the charger and 3.7v through my light. They definately make the Q-III brighter than it is with the regular 123s. About like it was with the stock circuit. I'll check later to see how hot it gets. Anything else I need to check?

I just did a heat check on my Q-III / Dorcy CX circuit / R123 setup. After 15 min the body had heated up to 102F and the star was at 113. After another 5 minutes the temps were actually a degree or two cooler. I'm going to assume the temps at 15 minutes were about max. The only problem I noticed was that the stock reflector, which had been taken out and put back in (so the glue bond was broken), became loose and rattled a little.

woolfam, if it's as bright as the stock circuit, then that's a good sign because the Dorcy circuit is supposed to be 18% less bright or less throw, I don't remember which. Hmmm... only 4.02v off the charger? What kind of R123s did you get? I got the cheap unprotected 750mAh (another post indicated that they're more inline with 650mAh cells, but considering the much lower cost, I'm not complaining) cells from Emilion on the BST non-lights forum come off the charger at 4.15v.

If you could, a rough runtime test with your R123 with the CX circuit would be helpful (is yours a stock Dorcy circuit or a modded one? e.g. short the resistor and add 10uF capacitor). But if you have an unprotected R123 like mine, it could be a bit tricky to test it without a fair amount of risk running the cell too low (below 3.0v when under load) and therefore killing the cell. But if you got protected cells, then run them until it cuts out, then re-test with the stock circuit. I would do the test myself, but I don't have protected R123s (nor a supply of primary CR123s), nor do I have the CX based circuit. I may purchase some protected R123s just so I can take out the guess factor when using the light, and not wondering, how much runtime have I had on this cell? Am I below the minimum cut off yet?

And heat my definitely be an issue with your light since you're running a J Vf star. On my K-bin star, the head gets warm after about 3 minutes of use. One weird thing with heat that I noticed though was that after I did the 2-stage switch mod, the whole body of the light gets warm, not just the head. (OT: I now know why the guys here charge as much as they do for the mod, it's not a simple drop in the switch and solder on a couple SMD resistors. The 3 legs on the 2-stage switch are too thick to fit through the holes on the circuit board, so they need to be filed/ground thinner, and then re-bent to the right angle, and then soldered in, but the added versatility is worth it).

My R123s are also the 750mAh ones from Emillion's web site. I have read that the voltage will increase after a few charges. We'll see... I'm with you as far as cost goes. I bought the Nano charger and battery combo and an additional battery for a total of less than $13 shipped. That seemed like a deal too good to pass up.

I have not modified the Dorcy circuit - I didn't know about that option. What advantage is there in mod'ing the circuit?

Since I don't have protected cells, I kinda hate to run them down too far.

I have often read on this forum that the heat-sinking capability of the stock Q-III is not good. However, my whole light got warm during the 20 minute run, so it seems to be dissipating the heat pretty well. I don't know if 113F is "too hot", but we'll run it for a while and see what happens.

Cost was exactly the reason I bought the same Nano + battery combo (I bought 3 cells in total) for the price of 3 Duracell 123s. And the only downside was the unprotected part, but that's why I bought 3 cells, so I can kill 1 and still be ahead of the game in terms of cost (as long as I get to use the cell 2 or 3 times).

At Vin=3.0v, shorting out the resistor (or removing it and connecting the remaining pads with a copper wire) will result in a 9% efficiency gain. The second tweak was to add 0805 or 1206 sized 10uF capacitor in the C3 position to add capacitance, which is supposed to help because of the circuit's "relatively low switching frequency", whatever that means. He didn't really quantify the benefit of adding the extra capacitor, but I'm guessing that the added capacitance may help smooth out any ripples in the current delivered by the circuit, sorta like a memory buffer for computers (that's my expertise). More details of this mod can be found in Doug S' thread that I linked to previously.

Looking at the design of the light, I also think it's not that ideal (the star dissipates almost all its heat to the upper edge of the aluminum ring, which isn't much surface area), but since the light does get pretty warm with use, it's doing it job. If you are really worried about heat, you can use a couple pennies and file them down to the right size and fit it into the Al ring for more surface area for heat dissipation like chimo did to his Q3 in a recent thread. My original plan was to make a "drop-in" pill using a modded Dorcy CX circuit, the Al ring, and an old copper circuit board for the positive contact, and epoxy the thing together (for good thermal transfer) with wires hanging out so that I could easily swap stars if I wanted to later on. But since I didn't get the CX circuit in either of my 2 Dorcys, I've been a tad reluctant to do that just yet.

[/ QUOTE ] (OT: I now know why the guys here charge as much as they do for the mod, it's not a simple drop in the switch and solder on a couple SMD resistors.

[/ QUOTE ]

Are you offering to do it for less? I've been thinking about getting one but I didnt really like the prices. or maybe you could just tell me how to do it? Send me a PM please

[ QUOTE ]
Are you offering to do it for less? I've been thinking about getting one but I didnt really like the prices. or maybe you could just tell me how to do it? Send me a PM please

[/ QUOTE ]

Just for the record, no, I am not offering to do the switch mod. It was a good learning experience for me (I did 2 for myself and 1 for my buddy), but I'll leave the mass modding to others, like Milkyspit, the originator. /ubbthreads/images/graemlins/wink.gif BTW, for a stock circuited Q3, stick with 2x 10ohm resistors in parallel like Milkyspit recommended (mine measured 5.6ohms). My 2nd mod I tried 2x 22ohm =11.6ohm (for when I finally swap in a Dorcy circuit), but on a stock Q3, 11.6ohm makes the light end up like an ArcAAA, except whiter. The 5.6ohm is easily twice as bright.

As usual, found this about a week too late to be of much use. Sorry guys. /ubbthreads/images/graemlins/frown.gif /ubbthreads/images/graemlins/duh2.gif

AvroArrow, thanks for the measurements and runtime info! Believe it or not, I don't even have an R123 version of the pimped-out Q3 for myself... but that's gonna change soon! Just got some little toys from Battery Station, heh heh. /ubbthreads/images/graemlins/naughty.gif

I should mention that the Dorcy circuit is Vf sensitive. (Did I say this somewhere already? Sorry if so!) What that means is the lower the Vf of your emitter, the more current the circuit will pump through it. Now it's possible a very low Vf combined with a fresh R123 might take things a little TOO far! /ubbthreads/images/graemlins/ooo.gif But in general, IMHO you want to go with a low Vf. The original pimped-out Q3 mods used SY0J stars, and they still do a great job... but lately my bin of choice is TYAH. /ubbthreads/images/graemlins/grinser2.gif

Hope they work with the R123! TYAH-Dorcy 1AAA-R123... now THAT would be a little monster of a light. /ubbthreads/images/graemlins/wink.gif