GDuP progress thread.

[LED Zeppelin]

Am interested new LE, as have quite a few "hosts". I do have a Ti Mule and am wondering if anyone has tried this LE with a reflector or is it only targeted for "mule"?

I would prefer a Aleph 2 head and reflector retrofit, but I see only A19 XRE reflectors?

Any insight would be greatly appreciated?

jeffb

jeff, I tried it out in a head with a McR27XR, really good throw as would be expected.

I just tried it in an A2 head that has a McR20 which I shaved off 0.030" for focus with the Seoul LE. The top of the Cree metal ring stopped at the bottom of the reflector, and the back of the LE was flush with the end of the head. It worked fine, but the focus was off. The hotspot was broad, and had some dark artifacts. The Cree needs to be raised more into the reflector, but this might separate it from ground contact with the tube, depending on how much more it was threaded into the head.

The stock McR20S has a larger emitter opening that the McR20 with 0.030" shaved. As Nate mentioned, the Cree ring may be able to raise up into the reflector instead of bottoming out. Then you could focus it better simply by screwing the LE in until the point it touches the leads and shorts. You could then add a wire ring between the tube and LE to make contact. Or as John suggested, shim down the reflector. It might very well work, but the beam might not be as good as the A19 head.

I'm glad jch79 tried it with a Kroll. I don't have a Kroll tube that would work but it's nice to know the GDuP works well with it. I have an ROF multi-function and a FluPIC that I've tried with Krolls. The ROF works great, the FluPIC skips from the contact bounce and is frustrating.

[dat2zip]

Good thing I held off ordering a proto run of new boards using the new IC. I had high hopes this would solve the low current sensing issue. Yes, the new IC runs on lower voltages and has better accuracy at lower sense voltages.

I guess I had hoped it would work well beyond it's datasheet specifications. What I need is an IC that will run down to 2V or so. (2.2V would be nice)

The new IC is only rated for supply voltage operations down to 2.7V and the one I was using was specified for 3V operation.

Both in this case operate very similar and stop working around 2.6V which is not low enough to get 1mA to the LED. Need something like ~2.3V or lower to be on the safe side and to work across all LEDs. I prototyped it up on a existing board and then ran it through a series of calibrations to see where it would drop out and stop working.

There is still the option to not use the sense IC and wire the sense resistor to the low side vs the high side. If it was moved to low side sense then we would not be able to make sandwiches which need common ground drivers.

This may still be a better solution to get the lower current ranges and wave off the sandwich compatibility option.



Wayne

[dat2zip]

A side note. I did the math again for the height requirements for the standard ESCan, ESink combo. Yes, it needs the shorter height inductor.

I went back to the site where I thought I had bought the inductors and spend more time redoing the research. After I chose an inductor that was close, but, left no margin (0.05mm, .020") I proceeded to checkout and when I logged in I found I still and the previous order still sitting there. The inductor that I really wanted quantity had dropped and I was not able to order the full quantity needed for a production run. I took all they had and got some of no margin inductors as well.

I'll have enough to make protos, but, someone would need to twist my arm to make Seoul LEs using this combo. Adding the anode board will have to be installed with zero air gap between the boards which is not an easy thing to do. It's too easy for the boards to get slightly cocked or pop up when you go to solder it in place.

The LED wires must come out and be near the mating holes on the EScrew or there won't be room to bend, twist or fold the wires. So, until it's tried I still don't know if it is even possible assuming the boards do fit in there.

I can not estimate any time line for the next protos. Due to uncertainty in design I would estimate the next target date to be late January.

Wayne

[Ms_SS]

The pre-production GDuP LEs have been price adjusted and are available now.

[McGizmo]

OK,
I am in the game now. I received some goodies from Wayne yesterday and one was a completed Aleph Mule with clickie pak and the GDuP. This one has the longer time out if I am getting the terms correct so a couple times I got into calibration mode but I was quickly able to figure out that I needed to hold a level a bit to "set" it. Having three levels at your finger tips is great! The only other light similar to this I have is the Draco and I think I prefer the momentary/ latching action of the clickie as a means for level selection over the twisty means.

Very cool Wayne!!

[alex in germany]

Hi,

so i can easily change the LE im my Aleph Mule fom McGizmo? Without soldering etc..?


Alex

[Ms_SS]

Yes, just swap out the NG light engine and screw in the GDuP LE.

[jch79]

I'll have enough to make protos, but, someone would need to twist my arm to make Seoul LEs using this combo. Adding the anode board will have to be installed with zero air gap between the boards which is not an easy thing to do. It's too easy for the boards to get slightly cocked or pop up when you go to solder it in place.

Wayne, to make sure I understand this point: due to the board thickness, making a GDuP with a Seoul LED will be difficult? After seeing the GDuP in action, I'm hoping to eventually upgrade all of my Aleph series LE's to GDuP - which includes a lot of Seoul/Lux-friendly lights.

Does this count as an arm twist?

john

[dat2zip]

Wayne, to make sure I understand this point: due to the board thickness, making a GDuP with a Seoul LED will be difficult? After seeing the GDuP in action, I'm hoping to eventually upgrade all of my Aleph series LE's to GDuP - which includes a lot of Seoul/Lux-friendly lights.

Does this count as an arm twist?

john

I've done the calculations for the board stack height to fit in the ESCAN. Whether or not the wires will fold,bend or exit correctly has yet to be determined since there will be little space from the top of the board to the EScrew. Since I have yet to order both boards in the proper thickness we can not try one now.

The more I think about this and the more I use it I'm totally shocked how long 1 second appears before the blip/timeout occurs.

Since cycling can occur as fast as possible it is probably better to set the timeout in the range of 1/4, 1/3 second.

While this may initially seem short I don't think most users will have any problems changing levels.

What this does mean is that timeout occurs at the moment you hesitate which would be the level you are going to stay at. By the time you start using the light, timeout would have occured and you would be using the light instead of thinking about when will the timeout occur.

This is very important in those moments when you use the light to quickly "spot" something. Is that a dollar bill behind the soda machine? Under these momentary spot situations I've found I can spot or get the information I need and have the light off in less than a second. What' amazing is how fast the human eye/brain gathers information. Under these spotting conditions the UI is actually in the way if the level is not the level you want when you turn the light on.

Ideally, you'd notice the light is not the correct level and you'd blip it to the correct level and immediatly spot your target and make a judgement turning the light on when satisfied.

Under these short timeout there would be no need for the blip visual indicator as it would not be needed.

Wayne

[LED Zeppelin]

Wayne,

I think the short timeout makes a lot of sense. The more I use the light I agree - no timeout blip, and a moment of hesitation latches the level.

I too find myself thinking about the timeout more than necessary.

[jch79]

Is that a dollar bill behind the soda machine?

No, it's just a lousy Snickers wrapper. All that work for nothin'!

What' amazing is how fast the human eye/brain gathers information.

Or in my case, it's amazing how slowly they gather it!

Joking aside, I agree - when I get to the desired level while cycling through the three levels, the decision to stick to that level is made almost instantaneously. So I'm with LedZep on this one as well!

john

[McGizmo]

I've done the calculations for the board stack height to fit in the ESCAN. Whether or not the wires will fold,bend or exit correctly has yet to be determined since there will be little space from the top of the board to the EScrew. Since I have yet to order both boards in the proper thickness we can not try one now.

The more I think about this and the more I use it I'm totally shocked how long 1 second appears before the blip/timeout occurs.

Since cycling can occur as fast as possible it is probably better to set the timeout in the range of 1/4, 1/3 second.

While this may initially seem short I don't think most users will have any problems changing levels.

What this does mean is that timeout occurs at the moment you hesitate which would be the level you are going to stay at. By the time you start using the light, timeout would have occured and you would be using the light instead of thinking about when will the timeout occur.

This is very important in those moments when you use the light to quickly "spot" something. Is that a dollar bill behind the soda machine? Under these momentary spot situations I've found I can spot or get the information I need and have the light off in less than a second. What' amazing is how fast the human eye/brain gathers information. Under these spotting conditions the UI is actually in the way if the level is not the level you want when you turn the light on.

Ideally, you'd notice the light is not the correct level and you'd blip it to the correct level and immediatly spot your target and make a judgement turning the light on when satisfied.

Under these short timeout there would be no need for the blip visual indicator as it would not be needed.

Wayne

This gets my total agreement! I played with one of these in the first iteration of slow time out in an Aleph Mule last night and I also put one with the quicker 1 second time out in a LunaSol 27.

IMHO, if a light is used primarily in a constant on mode then the time out is not such a big deal. If one uses a light with momentary activation often then time out is a very big deal and if you find yourself waiting for the light to lock into the desired level before turning the light back off, it is quite frustrating.

There is fantastic merit and value in having access to three levels of output and ultimately I suspect many would be willing to make concessions and accept work arounds to have these levels available. I think a much shorter lock out, lock in or time out or what ever it is termed would minimize the concession for a user like myself. I consider myself a momentary user and don't like a price placed on momentary use. Having the light cycle through levels as I go for momentary illumination is a price I would prefer not to pay.

In the case of the Aleph Mule, three is better than one and I will take the three regardless of price but still would want the light to be stable in output for momentary usage. (Short timeout) I would ideally have it so that I had to rapid fire the switch for level change.

The LunaSol27 is a different animal. For basic low, you have the flood of the nichias. For that momentary look beyond, you shift to the Cree. With the GDuP driving the Cree, you have access to three levels of the Cree and these all have their place and the output variations were quite cool to experience. If one were to use this light in constant on illumination then you simply can't beat the versatility of the GDuP. However, for me and my typical momentary only use of the Cree, I was frustrated having to wait on the Cree to lock prior to turning it off. I suspect that rapid time out would aleviate this issue if not minimize it.

At the present 1 second time out, for me, I would grab a single speed Cree LunaSol over the three speed version for typical EDC carry. If I were to grab the light with intent or need to have the Cree on for any significant duration, I would want the 3 speed.

To conclude, I believe for momentary users like myself or momentary use lights, I suspect rapid time out at 1/4 - 1/3 second would make a world of difference.

Working with this new GDuP was a treat and having three levels now available from the converter is a quantum leap in the magic contained in the small buck/boost driver. Hats off to Wayne once again!

On the LunaSol27, the GDuP initially gave unexpected and undesireable results in light output levels until I succesfully ran the converter through its calibration sequence. This in itself was quite facinating for this simpleton and most impressive!!

This is a cool driver and I know Wayne has a lot of time and resources invested in it. I hope that I can find ways of using this driver and take advantage of what it has to offer as well as support Wayne in his efforts in bringing this to us.

[dat2zip]

I would like to clarify something Don said. Don has one of the Beta raw aluminum lights that has no blip and the longer 1.5 second timeout. His sentence.

"In the case of the Aleph Mule, three is better than one and I will take the three regardless of price but still would want the light to be stable in output for momentary usage. (Short timeout) I would ideally have it so that I had to rapid fire the switch for level change."

is referencing the Aleph mule beta unit.


================================================== =======
I'm looking for feedback as to why no one is buying the GDuP light engines and lights now offered at the shoppe.

No communications is bad. I have not heard privately or publicaly from anyone one way or the other. The GDuP costs real tangible money on my end and many hours of time spent.

I believe many were turned away initally with the price. We have since adjusted the price to what we think is reasonable.

Is the price not in line with you desire?

Is the perception that you won't buy it because it's pre-production?

Unless I get feedback the GDuP will not be worked on. I don't hear or get any feedback even here on the current progress or posts suggesting which direction the GDuP should go forward as far as sandwich compatibility or gaining the low end LED current range.

I can't afford to order more parts for the next proto development cycle if I can't even recover the material and labor costs for just the protos that are now offered. There would be no reason to. The shoppe is not Amazon.com. The shoppe is Cindy and I. Our products and offering is specifically for CPF. Your support/contribution is through sales. Your orders are in essense your contribution of capital investment. Each CPF'er contributes to the development of new products.
================================================== =======

Speak up or let the GDuP die.

Wayne

[Nake]

I'm looking for feedback as to why no one is buying the GDuP light engines and lights now offered at the shoppe.Wayne

I'm waiting for the Turbo version. I would like the top light level to be at 1A.

[mfrey]

Same here. I'm interested in 1A or thereabouts, also, for the high.

[easilyled]

Sorry to hear that so much is at stake Wayne.

Speaking for myself, I was thinking that it might be wiser for me not to be
an early adopter, but to wait until further feedback about these boards had
come from other users and any additional refinements had been completed.

In addition, the majority of my Aleph lights are only compatible with Seoul
rather than Cree-XRE led's (and I was hoping that Seoul GDuP-LEs
would also be offered).

I very much appreciate all the time, effort and expense you have put into
furthering the development of these converters to give us great products
and I am sure that the GDuPs will really take off given a little time.

[dat2zip]

You guys will be waiting a long time.

Had you said something earlier I could have made some of these 1A version.

Something you now need to consider. Investment and funding. My investment has been many hours and many $$$. Your buying one LE for $100.00 or so whether you use it or not helps fund this development. That's $100.00 out of your pocket. My investment is the total lot. That's 10X-100X that or more.

If the development costs are not recovered then this will be the end of the road for this. I will consider alternative avenues and different methods to ensure I stay financially afloat.

Whether that be pre-pay, sign up lists or other means, if that is what it takes then that might be what I'll have to do.

There is nothing wrong with these LEs. There will be no more new, better or different ones if these don't move. If you are waiting, your wait will be forever. No production will be launched, new protos, nada.

I won't prototype or attempt a Seoul LE until this phase of the project is done and I've got too much investment to just toss it in the circular file and move forward.

I'll say this again. I've got all the boards I need for my personal use. If this configuration is not what you wanted you should have spoke up sooner. Too late now.

It seems there is a large number of silent viewers out there. If you don't express you interest here publicly or privately there is no way I can read your mind. I'm no magician.

If you don't want the LE, buy one of the Mule heads or completed lights with the GDuP in it. I'll move more LEs into completed lights if they start selling.

Wayne

[acourvil]

I'm very interested in the new drivers, but just ordered 3 sharks and need to finish those projects before I start on the next. Also, my preference would be to use a Q5 or R2 emitter instead of the Q4, or have the driver available separately from the LE.

[dat2zip]

I'm very interested in the new drivers, but just ordered 3 sharks and need to finish those projects before I start on the next. Also, my preference would be to use a Q5 or R2 emitter instead of the Q4, or have the driver available separately from the LE.

Your missing the point. Production units may be available in bare converter board, kits etc. But, these are not production yet. To request this now is absurb to say the least.

Wayne

[dat2zip]

This thread is almost becoming comical...

From the movie short circuit as Ben Jabituya would say.

"I am standing here beside myself."

Another great quote from him in the movie.

"I don't know about you, but I am planning to scream and run."

Wayne

[LED Zeppelin]

Wayne, dont' get discouraged.

I think it's just the timing with the holiday budget. Also the XR-E only configuration, and the limited compatibility with all the Aleph heads are contributing factors.

I do believe that if the 27L head/GDuP combo was ready you'd be pressed to keep them in stock.

The price is in line with the other Aleph LEs, which is very fair considering the versatility of this board.

If anyone's hesitating because of the "prototype" stage, I reiterate that it's good to go. The GDuP as it stands is not in need of revisions, and any changes to the production boards will surely be subtle. It works great and it's great to use. That comes from one who previously preferred a mechanical 2-stage Aleph 1, but now EDCs the GDuP Mule by choice. It's the perfect homeowner's light.

[Telkin]

I definitely have intesterest in the GDuP having spy'd on this thread without posting. However, one of the reasons holding me back from buying this driver is the learning curve. I'm not very technical to begin with and know very little about Alephs. I've glanced by this thread several times but I still do not know for certain how to change levels or even set levels (if that's possible). Also there has been changes to the driver since the original posts? It would be nice to get a summary of how things work.

I have an Aleph 2 host that i'm interested in using this LE but I dont' know what reflector I need to make that happen.

[LED Zeppelin]

Telkin,

The UI is super easy:

Click or press once to turn the light on.

Another press within 1 sec. changes the level (Lo - Med - Hi - repeat).

After 1 sec at a certain level, the light memorizes that level and stays. It will come on at the memorized level next time.

Once a level is memorized, change levels by pressing 2 times quickly. (First time it comes on at the same level, next press advances.)

To enter the auto-calibrate mode, press 16 times.

You cannot adjust the brightness of the different levels, they are preset. But they are well spaced and adjustment is not necessary.

The current offering of the GDuP LE will drop into your A2, but the focus will be off. For that head you could use a Seoul LED, and a McR20S reflector. Wayne will hopefully offer Seoul GDuP LEs for the A1, A2, and A3 heads. This LE will work perfectly in a Mule or A19, the latter is most similar in size to the A2 you have. The A19 uses the McR19XR reflector.

[Nake]

To enter the auto-calibrate mode, press 16 times.

What is this mode for? What does it calibrate?

[jch79]

Telkin-

This is the simplest multi-level driver I've ever used. The UI was intuitive after the first click.

john

[LED Zeppelin]

What is this mode for? What does it calibrate?

I believe it calibrates the drive current to the specific emitter. It will ramp the brightness up and down for five cycles, then pause at lo, med, and hi. After that, it ramps up and down again for five cycles, and returns to lo level. The entire calibration process takes about 1 min 40 sec.

[Telkin]

Thanks LED Zeppelin, your explanation was exactly what I was looking for thanks!

I actually do have an A19 head I can use but it doesn't seem to fit right with a CR2 body and the small flat tailcap. The proportions are all off...I'll try to figure out a solution thanks!

[jch79]

I actually do have an A19 head I can use but it doesn't seem to fit right with a CR2 body and the small flat tailcap. The proportions are all off...

Funny, I have that exact configuration for my front bike light, and I like how it looks! To each his own!

[iconoclast]

I'm looking for feedback as to why no one is buying the GDuP light engines and lights now offered at the shoppe.

I can only speak for myself, but the main reason I haven't scooped one of these up yet is holliday budget.
They look like awesome boards, but unfortunately no one on my gift list would appreciate them as much as I would myself. Early to mid January is probably when I'll get a chance to play with one. Looking forward to it though.

As a side note, the amperage range is fine with me and not interfering with my desire to obtain one of these. However if a ~1A version existed in the future, I'd pick up one of those as well.

[dat2zip]

I can only speak for myself, but the main reason I haven't scooped one of these up yet is holliday budget.
They look like awesome boards, but unfortunately no one on my gift list would appreciate them as much as I would myself. Early to mid January is probably when I'll get a chance to play with one. Looking forward to it though.

As a side note, the amperage range is fine with me and not interfering with my desire to obtain one of these. However if a ~1A version existed in the future, I'd pick up one of those as well.

Thanks for letting me know. I had thought some of you don't have the budget right now. Silence was killing me along with no sales was perplexing and confusing.

Wayne