Pioneer 12X Blu-Ray-- BDR-205

Thanks for doing this, i'll have the results in a moment.


Well, the different spread and of the data points does explain the crossing of the lines at the beginning of the efficiency plots.
But after that the behavior of your tested 8x is identical to all the others - it follows the same pattern....

It does not explain why your 12x's drop in efficiency so suddenly at the higher currents..

The weird peak with the first one probably was an error (altho it would be VERY useful to re-check that part, at least the start), but since the random 8x follows the same patterns as all the other tested 8x's afterwards, while the 12x's don't...

Something is definitelly different about them! :thinking:



Uploading pics....



Ok, so the bad news... The comparison 8x (called "Rnd 8x" for "Random 8x" in the plots) is the lowest in slope efficiency i have seen - lower than the 8x being tortured at this moment. But it has a much lower Vf than most other 8x's. Once this is taken into account, it becomes better than the first murder candidate..

The good news.. While the weird efficiency "peak" at the very start was mostly just a result of differently spread data points in the plots, as shown by the new 8x (it crosses the efficiency line of the lowest 8x in the plots at the start, like the 12x's did), the 8x later behaves normally and follows the exact same pattern as all the other 8x's. It's efficiency flattens out!

With the 12x's efficiency drops at the higher currents, and that's what doesn't make sense. I thought it was because of heat developed during plotting, but then it would show with this 8x, since it's a low efficiency one and creates much more heat than most other diodes at the same currents.

What makes 12x's different from the 8x's is not the peak, but the "hump" they display in the efficiency plot, and the fact that their slopes are not straight (if they were, there would be no "hump")...


They definitelly don't follow the same pattern as all the 8x's!



It suddenly seems quite possible that they are new diodes.
In fact, unless there were some horrible errors made during their plotting (which is very unlikelly, if not completelly impossible at this point), they most definitelly are not the same as 8x's!


The only thing that i don't like is that they don't make straight lines... I need to plot one in high resolution to see what's going on.


But heck, as long as they are rated for a higher power, who cares, right?
I just hope the bends are not the results of kinks. :undecided:


EDIT: Kinks are unlikelly, they would show up much more severelly... But it seems like with 12x's plotting will once again require a high resolution...

In any case, i am suddenly becoming interested in these diodes! I got Dave's 12x sled a few days ago and the little diode heatsink just arrived, so i can also do the sled efficiency comparison over the weekend... :yh:

daguin said:

I tested the Pioneer 8X diode this evening. I tested it the same way I tested the 12X's and then I tested it going up at 60mA intervals. The first set of data is replicating the 12X test. The second set is to see if the interval is causing the strange curve.

Replicating 12X test

mA -- mW -- VF
60 -- 32 -- 3.95
100 -- 85 -- 4.38
150 -- 150 -- 4.77
200 -- 213 -- 5.01
250 -- 278 -- 5.20
300 -- 342 -- 5.34


Standardized intervals
mA -- mW -- VF
60 -- 32 -- 3.95
120 -- 112 -- 4.55
180 -- 189 -- 4.92
240 -- 266 -- 5.17
300 -- 342 -- 5.34

If FrancoRob plots both sets of data using the same formula in Excel, we can see if the strange curve is an artifact of the spreadsheet (and my measurement schema) or the diode.

Peace,
dave

Click to expand...

According to these data, hereunder are the three plots:

Just to clarify. There was only one diode. It was a Pioneer BDR-203 8X. the data was simply derived in a different manner for each data set. I tested the diode the same way I did the 12X's AND in a way that would avoid what IgorT suggested might be problem with the "intervals" I was measuring.

PO and PI/PO SHOULD be the same. They are the same diode.

What we were checking for was the strange curve in the efficiency graph of the 12X. That does not seem to be there in either testing protocol.

There is a slight difference in the efficiency graph between the two tests, but nothing like the graph for the 12X. The efficiency graphs look like the ones for the other 8X's. They do not resemble the efficiency graphs of the 12X's

Hmmmmmmmmmmm . . . .curiouser and curiouser

Peace,
dave

If you like a comparison between IgorT Efficiency plot (Pi(mW)/Eff.%, hereunder is the Dave data equivalent plot:





Just to have a common protocol with Igor Efficiency plots (Eff% / Pi - mW), I'll modify all my Efficiency plots accordingly.

FrancoRob said:

Just to have a common protocol with Igor Efficiency plots (Eff% / Pi - mW), I'll modify all my Efficiency plots accordingly.

Click to expand...

The reason i plot Efficiency (%) against El. Input Power (Pe) is, that efficiency shows how much of the input power comes out as optical power.
Diode efficiency is directly related to electrical input power... If you read off the percentage at any point in the plot, and then calculate that percentage of the Electrical Power directly below that point, you get the optical output power of the diode...


Also, Franco: You mistook Daguins two different tests of one 8x diode for two different diodes, and marked one of them a 12x in the above plots..



Otherwise, i did another set of plots, to test one thing i was interested in...

Until now it was almost a rule, that a higher power diode would have a higher Ith and form a different plot. But i realized that GGWs and 8x's all make straight lines with efficiencies that overlap between many of them, and Ith's that are not much different on average... If it wasn't for the fact that 8x's can survive a higher power, slope efficiency plots would not really show a difference.

In this case, 12x's actually show more of a difference compared to 8x's than 8x's do compared to GGWs!


So something is definitelly different about these 12x's. And if they are different, they can only be higher power diodes...
Other indications of this were the sled itself - not only is it the same, but doesn't even use a beam circularizer to make better use of the diode's power!
The lack of the beam circularizer cube could mean the 12x's have a better aspect-ratio (= a rounder beam)...



Then i remembered, that there is one more thing, that can help us distinguish different power diodes from each other! It's their Vf!


No matter which 405nm diode we used so far, at the powers they could survive, they would have a Vf of around 5.5V. It would vary from diode to diode, but it would rarelly go above 6V unless a diode was severelly overdriven....

Basically, every time we get a new higher power diode, we set it at a higher current. But their Vf stays pretty much the same on average, as it did with lower power diodes at lower currents - it varies around 5.5V....

A higher power diode will have a lower Vf than a lower power diode when set to the same current. To reach the same Vf as with a lower power diode, the higher power diode will have to be set to a higher current...

As diode powers increase, currents go up, but the Vf stays the "same"...


So instead of trying to distinguish them by their slope efficiency or actual efficiency, which did show some differences, but still overlapped with many of the better 8x's, i made a new plot, comparing only their Vf's and ignoring all the rest.



This is what happens if i plot Vf's of 13 8x's and 2 12x's against current...
As before, some 8x's overlap, but much less of them than before, and on average the Vf of the 12x's is clearly lower than that of the 8x's!

The two tested 12x's ARE showing signs indicative of higher power diodes! :yh:


P.S. Not saying run out and buy them just yet.. They are still overlapping with two 8x's in Vf, and two tests can hardly tell us an average or proove anything, but the differences are suddenly becoming more and more apparent. Chances are becoming very good these actually are 300-320mW Pulsed diodes!


P.P.S. When viewing this plot, ignore the difference between 40 and 60mA, it simply comes from differently spread data points at currents where Vf changes rapidly.. What we are interested in is higher up. And there the two 12x's definitelly show the lowest Vf's!

Incidentally, 8x diode #1 and #2 which i plotted red and dark blue have the same Vf's as these 12x's, which i also plotted red and dark blue only with fatter lines. The two thin 8x lines poke out a bit towards the end of the plot.

Thanks for all the effort Igor!!! That's some really good data.

What you said about the forward voltage makes perfect sense. Your graph proves it too.

I would be willing to bet that these diodes are different.


However only 1 way to know for sure...

Please keep us updated on the 12x murder test !!!

Tabish said:

Thanks for all the effort Igor!!! That's some really good data.

What you said about the forward voltage makes perfect sense. Your graph proves it too.

I would be willing to bet that these diodes are different.


However only 1 way to know for sure...

Please keep us updated on the 12x murder test !!!

Click to expand...

You can be sure i will...


I'm just starting to wonder if the 8x's will die in time to start the 12x test. :crackup:

Not that i want them to, but... I might have to cycle the 12x together with the second 8x.. :angel:


Only problem is - i'd need another giant heatsink for that!

IgorT said:

..........
Also, Franco: You mistook Daguins two different tests of one 8x diode for two different diodes, and marked one of them a 12x in the above plots..
..........

Click to expand...

Sorry, but I was not willing to write a book in the plot log as the curves were together with the relevant data, but if this is cause of misunderstanding, I'll change the description of the curve, as per example: "Daguin 8x - rep. 12x test"....

FrancoRob said:

Sorry, but I was not willing to write a book in the plot log as the curves were together with the relevant data, but if this is cause of misunderstanding, I'll change the description of the curve, as per example: "Daguin 8x - rep. 12x test"....

Click to expand...

I think that there is enough info here in the thread to limit any misunderstanding.

Peace,
dave

I just received the Plextor PX-B940SA 12x Sled from bug373.

Wanted to take many pictures of it's dissection and so on, thought it deserved it's own thread like the Pioneer, but then i got this funny familiar feeling when i ripped off the 405nm diode's heatsink.

The feeling got even weirder when i started "undressing" the red diode. It was in two heatsinks and by the time i got to the inner one, the heatsink alone was barelly larger in diameter than a regular 5.6mm diode.. And from the top, through the tiny hole it looked like an open can.


So i pulled out the 12x sled Dave sent to me, put them side by side, and indeed, they were completelly identical!


Plextor is using the same exact sled as Pioneer to make their 12x drive....


I thought it would be interesting to test a different 300-320mW Pulsed diode, even tho i knew chances were good a different sled could contain the diode from the same manufacturer, as is the case with Pioneer and LG 8x's...


Well, if nothing else, i get to plot two 12x diodes from the same manufacturer, and i intend to do it in high resolution, since these don't seem to be too prone to producing straight lines when plotted, unlike 6x's and 8x's...


Anyway, here is a pic of the Plextor 12x sled. Anyone who has a Pioneer 8x sled around should recognize it:






P.S. One thing that surprises me is the huge distance between the 405nm diode and the first optical element in the sled...

At this distance, the cube can not possibly capture the diode's entire output!

8x sleds at least have a "beam circularizer" in between, to optimize the sled's usage of the diode's power.


This could mean one of two things (or both?):
1. 12x diodes have a MUCH better beam profile - a better aspect ratio, perhaps a narrower fast axis?
2. 12x diodes have so much power that it is not necessary to increase the yield of their output?!?

The aspect ratio will be easy to compare to other diodes, since it should show as a rounder beam..


I really need to test how 405nm diodes behave in this sled! Perhaps i could take a photo of the diode output hitting the cube if i used smoke... Luckily Dave also sent me an intact 12x heatsink, where i can put them in for the experiment... Which brings me to:


P.P.S. Dave, i can't believe i asked you to use the "squeeze-squeeze" method of diode extraction on this heatsink! I did not realize what i was asking! With this kind of a heatsink, i simply snap it in half, like we did with PHRs. It's the GGW and LG 8x heatsinks where the squeeze-squeeze method works wonders. That material is soft. This one is much more brittle.

And yet you succeeded! :bowdown:

I can see now why you said it wasn't easy tho... :angel:

IgorT said:

I just received the Plextor PX-B940SA 12x Sled from bug373.
Wanted to take many pictures of it's dissection and so on, thought it deserved it's own thread like the Pioneer, but then i got this funny familiar feeling when i ripped off the 405nm diode's heatsink.

The feeling got even weirder when i started "undressing" the red diode. It was in two heatsinks and by the time i got to the inner one, the heatsink alone was barelly larger in diameter than a regular 5.6mm diode.. And from the top, through the tiny hole it looked like an open can.
So i pulled out the 12x sled Dave sent to me, put them side by side, and indeed, they were completelly identical!
Plextor is using the same exact sled as Pioneer to make their 12x drive....
I thought it would be interesting to test a different 300-320mW Pulsed diode, even tho i knew chances were good a different sled could contain the diode from the same manufacturer, as is the case with Pioneer and LG 8x's...
Well, if nothing else, i get to plot two 12x diodes from the same manufacturer, and i intend to do it in high resolution, since these don't seem to be too prone to producing straight lines when plotted, unlike 6x's and 8x's...
Anyway, here is a pic of the Plextor 12x sled. Anyone who has a Pioneer 8x sled around should recognize it:

P.S. One thing that surprises me is the huge distance between the 405nm diode and the first optical element in the sled...
At this distance, the cube can not possibly capture the diode's entire output!
8x sleds at least have a "beam circularizer" in between, to optimize the sled's usage of the diode's power.
This could mean one of two things (or both?):
1. 12x diodes have a MUCH better beam profile - a better aspect ratio, perhaps a narrower fast axis?
2. 12x diodes have so much power that it is not necessary to increase the yield of their output?!?
The aspect ratio will be easy to compare to other diodes, since it should show as a rounder beam..
I really need to test how 405nm diodes behave in this sled! Perhaps i could take a photo of the diode output hitting the cube if i used smoke... Luckily Dave also sent me an intact 12x heatsink, where i can put them in for the experiment... Which brings me to:
P.P.S. Dave, i can't believe i asked you to use the "squeeze-squeeze" method of diode extraction on this heatsink! I did not realize what i was asking! With this kind of a heatsink, i simply snap it in half, like we did with PHRs. It's the GGW and LG 8x heatsinks where the squeeze-squeeze method works wonders. That material is soft. This one is much more brittle.
And yet you succeeded! :bowdown:
I can see now why you said it wasn't easy tho... :angel:

Click to expand...

Plextor is simply a "re-brander" now. It was really just a question of "who's" drive they were re-branding. Now we know it is the Pioneer.

I hope the beam profiles are different. In any event, with this one and what I sent you will have a couple of Pioneer 12X's to look at and one to kill.


I just pulled another diode from this heat sink. The member wanted to try putting a lower power diode back into the drive to see if he could make a reader out of it. I do NOT want to make a habit of that. It is a pain (and very time consuming).

Peace,
dave

I just plotted the Plextor 12x diode in high resolution, and i am more than a little surprised by the results!


I expected two things: (from Dave's previous experiments)
1. Lower Vf than with 8x's
2. Strange bends in the PI line


But instead i got the exact opposite! :thinking:


Here is the PIV Plot of the Plextor 12x diode:




The PI plot is no more and no less linear than with 8x diodes. It's not a completelly straight line, but some variances are only to be expected, and some of it could be due to small errors and heat developed during testing...

The Vf on the other hand is higher than with most 8x diodes i tested. Only a few had a Vf this high!


Until now, i thought the 12x's were different from 8x's for two reasons - lower Vf and less linear PI plots (and higher slope efficiency, altho that could have been random luck)..

But this diode seems very much like a (very) high efficiency 8x diode....
If it wasn't for one "small" difference.. It's Ith is way lower than that of any 8x i tested! The Ith of this diode is only 26mA!


So even tho this diode does not exhibit the strange behavior found during Dave's testing, it does show differences from 8x's.

One important thing is, that it's Ith is significantly lower than with any 8x diode, and the other interesting thing is, that this is the third 12x diode tested (to my knowledge) and AGAIN we see an efficiency WAY higher than with the majority of 8x's...

The Vf is strangelly high with this one, but it could simply vary just as much as it does with the 8x's...


Only thing i don't understand is why Dave found the strange non-linear behavior in the first two tested 12x's.

I still haven't received the 12x Murder candidate, i'll know much more then, but what are the chances, that Plextor would use an IDENTICAL sled, but a different diode?

It would be easy to discard the differences due to different testing methods or errors, but Dave's 8x comparison plot did not exhibit the strange 12x behavior, and that should have eliminated the likelyhood of errors! :thinking:



It's strange and i can't find an explanation until i test the Pioneer 12x Murder candidate, but even with the unusual differences, two things are for sure:
1. 12x diodes have lower Ith's than 8x's (i noticed this already when extrapolating Ith from Dave's data - couldn't get it above 29mA even when i tried hard (during my "correction" attempts), by itself it wanted to stay at around 27mA, but i'll have to actually measure it)...
2. 12x diodes seem to have a VERY high efficiency compared to 8x's - only very few freak 8x's exhibit efficiency as high as these 12x's have almost by default, but even those 8x's that can keep up in terms of efficiency have a higher Ith... The 12x's do overlap partially with 8x's in terms of efficiency, but they clearly form a new separate higher efficiency group.



There are too many questions left open and speculating at this point wouldn't make much sense, i need to wait for the Pioneer 12x to get here (and i have to say i'm really anxious after this test!).

But i do believe these are completelly new diodes, different from 8x's...



P.S. The second attachment shows a comparison to a high efficiency 8x diode - you can see the Ith difference clearly there... This 12x has a noticably higher efficiency, but i think there was an 8x that gets close in the batch of 13 i tested.

I need to add the data from the other two 12x's, but later.... Right now i need to get back to work. But i think this one is a freak even compared to 12x's.. I'll check as soon as i find some time (shouldn't even have done this yet, should have been working, but i couldn't stand looking at it just laying around :angel..



EDIT: I took a look at the plots of the first two 12x diodes, and indeed, bug373 is lucky, he got a freak 12x in his Plextor drive! The power at 350mA is approx 20mW higher than that of Dave's higher efficiency 12x..

Up to the second one, the very high efficiency compared to 8x's could have been pure luck - a freak coincidence, but with the third one being even higher, i think it's safe to say these are completelly new diodes!



Otherwise, i should perhaps mention that i didn't actually use the 405-G-1 lens in this 12x test, but one of my custom lenses (the "Nichia Lens"), mounted in a metal lens nut by Jayrob.

I only have one 405-G-1 here, and it's occupied with the first 8x Murder candidate, but Jay told me that the "Nichia Lens" tested at the same power as the 405-G-1, and my tests confirmed this. If there is a difference it's less than 0.5%, so i simply used that, but i wrote "405-G-1" in the above plots, so as to avoid confusion and since it's easier for people who have one to relate to....

I wanted to do an efficiency comparison plot earlier as well, to see those strange "humps" 12x's form (or so i thought), but i was low on time. I also thought it was pointless since the P/I plot was straight.

But i need a break from work now and then, so here goes...


Observe what a peculiar thing happens when a Po/Pe comparison plot is made:



First of all, since this 12x diode has such a high Vf, once it is taken into account, and an actual efficiency slope is plotted, the two diodes practically overlap!


Only at the beginning is there a difference, because the 12x starts lasing so much sooner.
However, had this particular 8x been plotted up to 360mA, it would actually come out higher in actual efficiency at higher currents!

You can see this even better in the Efficiency plots, where the 8x starts lower, but suddenly speeds past the 12x at around 1.2W electrical power into the diode..


Interestingly, even in this higher resolution plot, the 12x still forms a tiny little "hump" at the start (unlike the 8x), as well as keeps that peak - both characteristics it exhibited in the low resolution plots.

This initial efficiency peak comes from the fact that it starts lasing much sooner than the 8x. As for the "hump", in the hi-res plots it looks so insignificant, that it's probably meaningless... All in all, the 12x actually makes a nicer efficiency curve, while the 8x is a little "jumpy" at the start. No idea where that comes from. Could be measurement errors once again (they are more pronounced with lower numbers), but smooth curves and lines somehow feel "healthier".. Don't know why (well due to the bad experience with kinks i guess)...



On average however, these two diodes have practically the same efficiency... Actual efficiency of course (how much of the electrical power going IN comes OUT as light (mW OUT per mW IN))..

In slope efficiency (mW OUT per mA IN) the 12x's still exceede the vast majority of 8x's (all but the freaks)...


They also show several other characteristics completelly different from 8x's, so it's becoming quite conclusive that these ARE new higher power diodes, which is happy news for everyone... :yh:



EDIT: Even more so, since they don't show those strange bends in the P/I Plots, which i feared might be kinks, afterall!
That thought really scared me for a while, because it could have meant, that they might not be so forgiving when overdriven...

Man! 460mW's @ 300mA's...

That's one 'HOT' diode! But over 6 volts forward voltage. It sure does seem like a different diode like you said...

..the high forwardvoltage could be a problem soon.. once we find out they are good for obscene outputs, and the (current?) flexdrive is not able to put out that high current at those voltages..

we need to quickly "finish" the 8x experiment, so we know at what level to test the 12x?

manuel

jayrob said:

Man! 460mW's @ 300mA's... That's one 'HOT' diode!

Click to expand...

Indeed, it's a proper freak, even for a 12x (altho we might see even more surprises as we test more of them). Attached is a comparison plot of the first three 12x's...

But believe it or not, there was one 8x in the batch that actually did 460mW at 300mA as well! And at 360mA it would have exceeded this 12x by 7-8mW (extrapolated)!

The plots would cross, because it starts later - at 29mA - but it's slope efficiency was actually higher..
That's one super-freak 8x! :bowdown:


But it stood far appart from the main group, so it's safe to say that 8x's like that are extremelly rare, while this kind of high efficiency seems common place with 12x's by now... With the third being even higher than the first two, it can't be a coincidence anymore.


Also, i just checked all the 8x Ith's, and the lowest i have found among them was 28-29mA (again, very rare, usually 31-36mA). The 12x's display their own Ith characteristics.


Some of these differences may seem small by themselves, but when all of them are taken into account, they become very significant!
Enough to say these are new diodes for sure.

jayrob said:

But over 6 volts forward voltage. It sure does seem like a different diode like you said...

Click to expand...

Yes, however this part is something that didn't quite fit in.

The Vf should have been lower than with most 8x's, not higher!
Well it has to vary a little from diode to diode, but it was way too far off to make sense... I couldn't stop thinking about it, i think i found a possible explanation...


I found something, that could have introduced additional resistance into the testing setup, at a point, where current flows through. This resistance would have caused a voltage drop, proportionate to the amount of current flowing through it (U = R x I -> the higher the current, the higher the voltage drop) which would make the Vf seem higher AND climb faster as a result.



After comparing the data to 8x's and 12x's (the values as well as the curvature and rate of climb of the Vf plot), i am pretty sure the above posted Vf plot is wrong!

I will need to repeat the plotting for the Vf data alone (the slope is correct, only the voltage was affected), to make sure we have correct numbers. I don't want it skewing the 12x data we have...


I will post corrected plots as soon as i have some time.
Until then the Vf part of the Plextor 12x plots should be ignored..

The Vf should have been much lower, and the actual efficiency of the diode MUCH higher than what my first test seemed to indicate.