Running the XJ-A140 With Missing Diodes

[kobra000]

i see you jut landed each side of the string to 3 or 4 planar/plate leds..i had thought about doing that myself,but with much higher power leds (30-36v 1.5-3amp, around 2500 to 6000 lumens of 450nm-470nm blue.all based on their being 32v @1.5 amp to each string..) and still trying to think of a way i can dimm it down to around3500 4k lumens without disrupting the voltage/current and having it shutoff all the time..wow you used diodes?..interesting..gonna have to read a tutorial about them cause i honestly dont know much about them.13 volts? why so low? ..its been said that each string uses puts out 32volts/~1.5-1.8 amps...why did you choose to use 13 volts instead of maxing it out or getting as close as possable 24v 28v 30v? for some reason i cant get a voltage/current reading cause mine wont get passed its "temp" error when i install a cable,and block the door interlock down and try to test it...just sits and blinks and shows nothing going to the Flexcable (KasEO cable)

would you have any pictures of your project? id love to see them.that would be awesome..I was beginning to give u hope alittle.thanks for posting man,you have refueled my nerd fire for these projectors=)

Keith

[rhd]

i see you jut landed each side of the string to 3 or 4 planar/plate leds..i had thought about doing that myself,but with much higher power leds (30-36v 1.5-3amp, around 2500 to 6000 lumens of 450nm-470nm blue.all based on their being 32v @1.5 amp to each string..) and still trying to think of a way i can dimm it down to around3500 4k lumens without disrupting the voltage/current and having it shutoff all the time..wow you used diodes?..interesting..gonna have to read a tutorial about them cause i honestly dont know much about them.13 volts? why so low? ..its been said that each string uses puts out 32volts/~1.5-1.8 amps...why did you choose to use 13 volts instead of maxing it out or getting as close as possable 24v 28v 30v? for some reason i cant get a voltage/current reading cause mine wont get passed its "temp" error when i install a cable,and block the door interlock down and try to test it...just sits and blinks and shows nothing going to the Flexcable (KasEO cable)

would you have any pictures of your project? id love to see them.that would be awesome..I was beginning to give u hope alittle.thanks for posting man,you have refueled my nerd fire for these projectors=)

Keith

- The projector is certainly not powering each channel with 1.5 - 3 amps. I believe the early model was powering these LDs at somewhere around 500mA per channel. The higher lumen projectors probably do more. I would be somewhat surprised however, if that figure was above 1A, as the Nichia spec sheet (until very recently) listed 1A as the max current for the LDs that we presume to be in these machines.

- Why diodes? What better to simulate the voltage drop of laser diodes, than more diodes? We're basically swapping out 6x LDs for 1x LED. The total voltage drop in those two scenarios is different. Diodes just seemed the obvious approach to evening it out.

- Why 13 volts? Well, largely because that was what I could find that would fit 4x emitters to the stock heatsink (almost perfectly actually). If you use a 32 V emitter plate that is meant to be driven at 2A, and drive it at 500 to 800 mA, you're probably not going to see much more output than you would from a 13V emitter plate that is meant to be driven at 800mA, and is actually driven at 800mA. Even LEDs have IV curves just like our LDs. If you really wanted to use all of that voltage, you could series two 13V 800mA plates, and you wouldn't even need to do anything else to trick the projector into thinking the channel was full of stock LDs. However, you will not fit 8x emitter plates on the stock heatsink (nor inside the unit at all, even with a custom heatsink).

Again though, truly, the only remaining issue is collimation of the LEDs. I've tried every short focal lenth lens that I could order from SurplusShed, none of them are able to get enough blue focused to a great enough degree to create any meaningful blue/green output.

[kobra000]

yeah i see what your saying...so have you put a meter on yours to get some hard core accurate data?...cause man i cant for the life me...the guys here t PL are as good as gold,but i dont know perhaps im over looking something or not understanding something properly...i still dont know why my PJ is throwing a temp alarm when its practically ice cold...got 5 xja-140's and each one throws a temp alarm when i got to test...was thinking of perhaps heating up the flexpcb and seeing if that would trick the thermal sensors into thinking that the heat sink is warming up..(though i still gotta get the heatsinks)..does the forward voltage loss of the diodes impact it very much? or are you using diodes with the Vf so low that its miniscule?. i got some 20ohm 30watt resistors the other day.but i havnt installed them,cause i was playing around with the idea of variable resistors and rheostats.basically one on each string (kind of a 'dial a load' type thing,but i dont think its going to go anywhere..that is unless your counting 8 of the individual "tendrils" as a string unto itself..i was counting every 2 as a string..figured one side was negative and the other positive..man there is a wide learning curve associated with these PJ's i swear.

what are the lumens outputs of your 13v led? at 800ma.?...gotta be under a 500 i would say..have you though about using higher power ones 18 to 20v perhaps. (matching it to whatever the actual power output of the strings is)...also...are you shooting the leds at the mirrors inside? and then using them as there stock function to bounce the light into the dichos?...i had something different in mine...i ll have to draw it out cause my descriptive skills suck.

thanks for posting man,really appreciate it.


thanks.

Keith

[rhd]

yeah i see what your saying...so have you put a meter on yours to get some hard core accurate data?...cause man i cant for the life me...the guys here t PL are as good as gold,but i dont know perhaps im over looking something or not understanding something properly...i still dont know why my PJ is throwing a temp alarm when its practically ice cold...got 5 xja-140's and each one throws a temp alarm when i got to test...was thinking of perhaps heating up the flexpcb and seeing if that would trick the thermal sensors into thinking that the heat sink is warming up..(though i still gotta get the heatsinks)..does the forward voltage loss of the diodes impact it very much? or are you using diodes with the Vf so low that its miniscule?. i got some 20ohm 30watt resistors the other day.but i havnt installed them,cause i was playing around with the idea of variable resistors and rheostats.basically one on each string (kind of a 'dial a load' type thing,but i dont think its going to go anywhere..that is unless your counting 8 of the individual "tendrils" as a string unto itself..i was counting every 2 as a string..figured one side was negative and the other positive..man there is a wide learning curve associated with these PJ's i swear.

what are the lumens outputs of your 13v led? at 800ma.?...gotta be under a 500 i would say..have you though about using higher power ones 18 to 20v perhaps. (matching it to whatever the actual power output of the strings is)...also...are you shooting the leds at the mirrors inside? and then using them as there stock function to bounce the light into the dichos?...i had something different in mine...i ll have to draw it out cause my descriptive skills suck.Keith

The forward voltage loss (drop) of the diodes is the POINT of their employ. You need to drop the difference between what the projector expects (roughly 6 x 4.75 = 28.5V) and what the LED is itself dropping. In my experience, with the 13V LEDs, you need to drop around an additional 10V with diodes.

I've tried the resistor approach mentioned in the OP, with 25W resistors. It worked, but only for 4 or 5 minutes. The resistors eventually got so hot that they desoldered themselves from my setup. That said, they weren't heatsinked at the time. I just don't like the resistor approach, and find it more logical to drop the additional voltage with a series of diodes, each of which is handling a bit of the heat dissipation.

If you're getting a temperature alarm right from the get-go, then it's not a temperature issue at all. It's either:
- You're missing the thermistors entirely (or are not properly bypassing them)
- The voltage drop of one or more of the 4 channels is not what the projector expects
- The bottom cover is open, and you haven't bypassed the clicky switch

I'll try to attach a video. If it works, you can see one of my test setups running 4 blue emitter plates, one from each channel, and staple RED output. The challenge, is that with the heatsink and LEDs inside the unit, not nearly enough blue actually enters the optics to do anything meaningful for the BLUE/GREEN. As you can see, there is MORE than enough blue output from the LEDs. The colour floods the room, it's overwhelming. The issue is getting it collimated. Even in this clip, there are short focal length coated glass lenses on each LED emitter.

Sorry for the shakiness of this video, I was trying to hold my phone while operating this all.



EDIT: In response to your question about metering, no, I haven't tried. In my mind there's not a lot of uncertainty here. We know what the stock LDs drop in terms of voltage, and we know they're wired in series. So we know the total voltage drop of each channel. We also have a rough idea of what 445 LDs are run at in terms of current. My understanding is that all of the projectors we're using are running them at under 1A per channel (and per diode, since a channel is 6x LDs in series).

[kobra000]

If you're getting a temperature alarm right from the get-go, then it's not a temperature issue at all. It's either:
- You're missing the thermistors entirely (or are not properly bypassing them)
- The voltage drop of one or more of the 4 channels is not what the projector expects
- The bottom cover is open, and you haven't bypassed the clicky switch

1.nope got them all..i have 20 cables to choose from (thanks krazer=) of both the lpf25s and 20s (the 20's just seem longer)...and i use the best ones out of the bundle for testing purposes
2. the Flexcables are bare.no lasers nothing no leds connected to them,im just connecting it to the port then taping it to the outside of the projector,then screwing the bulb bay door down as it should be to depress the switch thats in there..
3. yes i have not bypassed any of the thermal sensors...i know all i have to do i short out the big white one with the 3 pin connector.but not sure how to do the other ones...those 2 real small ones on the right and left side and the or 4 real small ones mounted above each string/channel.

i like that little catch all photograph you posted is it ok if i copy and use it?...(is there a color code to it?) could you elaborate on its function just a bit more....can it be used as a hook up schematic for diodes,resistors, leds?
im really digging your diode method. does it run cooler than the resistor method? .tripped me out when you said yours was de soldering itself..i had the idea of using this http://www.ebay.com/itm/390222033103...84.m1438.l2649 and instead of doing many resistors just use one big one on perhaps 2 or 3 of the 4 strings. but i was told that with wire wound resistors and the fact that the PSU to this projo pulses at either 50 or 400mgz (sorry forgot which) that i would get some bad EMI on just about everything in my home.
i see what your saying about voltage now...just let the laser diodes specs be our data and go off that,but im curious as to how much room is there for error or modification?..,what if i wanted to us an 18,25,36v led and 1.5 to 3.8amps how far from 13v can i stray? (13v per stirng right?) since the LD's are ran in series..can a series cirucit be combined with another series circuit?...ie can strings 1 and 2 be combined in series to give more voltage?
also how many lumens is your 13v led putting out? see i was only going to use one very high output (in lumens) led something like this.see below


this one is pretty close in NM to the stock NM of the laser diodes,but doesn't have the lumen output that i want.
http://www.ebay.com/itm/280744793420...84.m1438.l2649

this one had real low current which i thought might be beneficialhttp://www.ebay.com/itm/360394220444...84.m1438.l2649

this one is a bit lower on its voltage but has a higher current (3.8amps) and hewwwwwww doggy does it pump out the light ( i know the listing is misleading,6,000/8000 .i have contacted the seller and told them about their typo and they assure me of the LED pumping putting out 8,000 lumens.)
http://www.ebay.com/itm/Blue-High-po...item3f0e676669


I was going to use one of these posted (still dont know which one) and then cut a custom triangular style heat sink...ill include a drawing when i can get my CAD to work...and mount it close to the lens that the lasers were shot through. i figure this way i wont get as much scatter loss.and at between 2500 and 8000 lumens....well im very curious to see what will happen.lol.
You should really check out anchor optics.com...they have TONS of stuff..( their catalog id 4 inches thick!).all kinds of lenses/filters.laser stuff.all kinds,.they might have just what you/we need..perhaps tearing apart a pair of binoculars or a rifle scope by have a lens you can use. perhaps and old camera or overhead projector lens assembly.

im really sorry if i tore your ear up with the questions..im just really excited to see some results on someones mods and very eager to see if i can apply them to mine..

thanks very much

Keith.

[rhd]

Hey,

1) 20 cables? You mean 20 flex cables?

2) So how did you expect NOT to get a thermal alarm? As mentioned, thermal alarm doesn't always imply thermal problems. They just presumably use the "Temp" light to catch all errors. That includes your error, which is coming from the fact that you haven't tricked the projector's "am I ok?" circuitry into thinking it has any LDs anymore. Have you read the OP? It goes over this all (that's what I read a few months back before starting my experimentation)

3) I've never come across the "big white one". None of my flexcables (A130, A140, A145) have anything other than the tiny black thermistors.

--

A) That design won't really help you I don't think. I'm not sure I understand how you would use a graphic of a PCB that doesn't have any part info, etc. It was definitely not a "production" or "user friendly" layout. It was really just for my purposes, as it achieved a pretty narrow objective. But if you have a way to use it, by all means, go nuts. I did post it out in the open after all

B) Using a 100 ohm resistor? Do you have some rationale for thinking that 100 ohm is more appropriate than the 15 ohm tested in the OP? Regarding applying it to all strings at once, do you have some rationale for thinking that the strings can be paralleled without throwing off an alarm with the projector's "am I ok?" circuitry? And for that matter, if it worked, wouldn't you want a 100W resistor since the OP calls for ~25W of heat dissipation per channel? You're really not selling me here.

C) How far can you stray from the projector's expected output per channel? I really don't understand the question. You can't get it to output more current/voltage than it intends to use in powering the LDs....so 36V abd 3.8 Amps are a no-go. You can easily use less voltage if you'd like. I'm using 13V. You could use 1V if you wanted... as long as you're dropping the rest with diodes or an appropriate resistor.

You posted a link to this:
http://www.ebay.com/itm/280744793420...ht_5236wt_1163
That requires both MORE voltage, and MORE current than a channel provides. It's also 58mm x 40mm. How would you fit that on the heatsink inside the projector ?!?

You also posted:
"this one had real low current which i thought might be beneficial http://www.ebay.com/itm/360394220444?ssPageName=STRK:MEWAX:IT&_trksid=p398 4.m1438.l2649#ht_2957wt_1163"
Again, that's NOT low current. That's higher current that a single channel provides. And while it doesn't provide dimensions, I think you'll find the same problem with fit.

And:
http://www.ebay.com/itm/Blue-High-po...ht_4536wt_1396
How are you going to supply 3.6A ? I'm also fairly certian that this particular emitter is wider than the entire body of the projector.

I think you're thinking along the ride line with respect to mounting the emitter close to the actual lens that the knife edge assembly projects through. I think that's the only way to do it.

[kobra000]

1.yes i have 20 individual flex cables,ten of the LSA25'S and ten of the LSA20's... (Thanks again to Krazer for hooking me up I appreciate it man=)..) all of them are good and complete less the 24laser diodes.

2.yes i have read this entire forum many times.simply put i figured it shouldn't throw a temp error because it wasn't hot...i didn't anticipate that they would run other errors through the "temp light",..

3.yes on the Flexpcb there is a white 3 pin connector on the left hand side mounted in the corner (see the reverse side of the white sticker that says LSA-25)..i was told it was thermistor (1.5"long 1/4"w 1/8"thick (approx) with 2 wires w/heat shrink tubing connect to this...A forum member ,DanielBriggs, (another great guy by the way ,will hook you up=)....) told me this about it .
[QUOTE]""To answer your question: the little 2-wire thing is a thermistor; they are thermally adhered to the aluminum diode block, and MUST be connected for the projector to boot/operate. They monitor the diode block temperature (it gets really really hot!) and the projector shuts down / goes into lockout mode when the resistance goes too high (or disconnected.)
I never investigated the spec of them; however I'm sure similar thermistors could easily be sourced should you need many more to get projectors back and working.to defeat it,short the pins together,its just looking for continuity.[QUOTE]

but i have not defeated it as of yet,nor have i defeated the 2 on the right and left side of the pcb or the 4 that are at the top of each string.since they are so small,im not really sure how to defeat them.

A: ok im official confused on what the graphic of the PCB is for?...i thought it was your home brew schematic on how you wired in your diodes? it shows EM1+ and EM2- so i assumed it was your LED wiring/diode wiring method.


B yes the op suggested a 15 ohm resistors,but 15ohmx6 equals 90ohms ( which is i assume the total resistance of an entire channel. figured there might be as much as a 10ohm loss because of heat so i compensated)....whats the difference of using many resistors and using one big resistor? ( the op said one could possibly use a big resistor to replace a whole string) so instead of soldering a bunch of smaller resistors to the flexcable in place of LD's...i figured a larger one could do the same job per string. But then i started reading more about resistance and resistors and whatnot then got to thinking about Variable resistors cause i had read that they can be used to find an unknown resistance (read it here http://physics.kenyon.edu/EarlyAppar...surements.html )and figured one of them could give me the precise resistance/ohms that makes the projo happy and turn on. and was told that (assuming Krazers/OP theory of 32v@1.5a per string) that the only thing that could handle that type of load was a Rheostat (just another name for a variable resistor but one that can handle high power/current) but i have almost given up on this idea because the only variable resistors/rheostats i can find that are suitable are the size of my hand.hence impractical to use (unless i wanted to go external and make some sort of decade box/external resistor box,but dont want to have to lug something like that around) and the large resistor i had previously posted is capable of handling 50 watts,so i figured it was more than suitable.except for the whole EMI (Electromagnetic interference) and inductance issue. so now im looking into snap ferrite EMI suppression cores and also EMI shielding tape.

Do you have a part number on the Diodes you used..id like to take a look at them.

C:what im failing to see is if your saying the entire laser array only uses 28.5 volts @ 1 amp (1 amp for all 24 lasers) or if your saying each string uses 28.5 @1amp (which should give me a theoretically total of 114v @ 4 amps)
if each string is ran in series and each string is producing 28.5 volt,then it should be entirely possible for me to connect 1 string to another,(series circuit to series circuit) get more voltage and current,use whatever i need to power what i want ,and then feed the surplus to diodes and or resistors to bring the entire load down.
Ok what im interpreting is this,out of the available 28.5 volts in one string your using 13v and feeding the rest to diodes to (not entire clear on your hookup method) .now if its confirmed that the total output of 1 string is 28.5v (4.75vx6=28.5v)..then i should be able to use up to and including 28.5 volts from 1 string (six potions) to power whatever led i choose.from a 1 volt to a 24 volt led (i have a xj-a 140 @ 2300 lumens so im going to assume its got more current.) @1 amp current
.
Ive enclosed a picture of what i mean.. please not that the polarity positions marked in the picture/drawing are approximated.(havnt found anything that tells me which trace on the flexpcb/KaseoCable is + or -)

I Was planning to mount the led horizontally instead of vertically inside a 2 part custom cut aluminum heat sink and use a prism to bounce the light@90 degrees into the primary lens (where the LD'S light was shot through) i figured this would give me much more surface area for heat distribution while keeping the over all heat sink within stock size.
see attached picture:
sorry for its quality,im right handed and have nerve damage in it so im not to accurate with drawing.

yeah so thats my take on it all....i know it sounds fubar but it clicks in my brain.

Keith.,

ps for some reason its not displaying my picture of the Flexcable very well...probably have to download it and zoom in to see my notes. Please pay no attention to the attached thumbnail for so reason i cant get rid of it for some reason.
also here is a un alterned picture of the Flexpcb cable if you wish to use it to draw out what you have dont and annote it..(think that might give us a better picture of your setup)

[kobra000]

here is a really interesting .pdf i found on anchoroptics.com covers offers very straight forward stuff on all things projector concerned.thought it might be useful to everyobod
http://www.anchoroptics.com/document...load.cfm?id=16
also a thing on collimation systems
http://www.anchoroptics.com/document...load.cfm?id=14
and the link to their entire download library..which has many many projects covering everything from building opague projectors to 3d stereoscop viewers..just lots of stuff
http://www.anchoroptics.com/documents/

hope you guys find it informative.

Keith.

[kobra000]

this is fantastic comparison of all kinds of lights taking through many different levels....makes me start thinking about the Cree XM-L instead of the 50watt planar LED.

.

[kobra000]

this is a great run down on at least 30 different types of Leds,specs below each one......great list.
http://forum.lowyat.net/topic/609074

also ebay has a good lot of Collimators and whatnot

also check out edmund optics,they have lenses of every single configuration you can dream of.