Strange Artifacts of new LPM's output...

[lasersbee]

When I saw the detailed Review and posted power Graphs
of the Rubicon here...

http://laserpointerforums.com/f52/first-review-delta-rubicon-lpm-84268.html

I had noticed a few things that jumped out at me that didn't
look quite right but I had no time to investigate at that time.

In the past few days I took the time to have a look at it and
did some basic tests.

RINGING

If you look at the beginning of the graph you can see an
oscillation of the signal for ~4 seconds in one graph and
~6 seconds for the other graph before it stabilizes.

It is commonly known as ringing.

Notice also the maximum swing peak to peak is about
210mV (or mW) on one Graph and ~75mV on the other
Graph.

(For the sake of this Post and since we are using an OPHIR
20C head mV and mW are interchangeable)

If you compare that to the Graph supplied by BLORD of the
same Laser the beginning of the Graph does not seem to
exhibit the same ringing artifact.

I wondered if the ringing was actually coming from the OPHIR
heads and the fast sampling rate of the Rubicon was actually
picking up this previously unseen ringing.

I decided to hook a new pristine OPHIR 20C head powered by
two freshly charged 9V batteries to eliminate any possible PS
noise and see the output on our 100Mhz Digital Storage Oscilloscope.
This DSO can sample the input much faster than any hobbyist's LPM
firmware to date and should able to pick up any ringing artifcacts.

Here are some pics of what the scope showed... Remember that
these waveforms are raw output of the OPHIR head directly into
the DSO and have no additional filtering.


(DSO Full swing)

(DSO Overshoot)

Here is a shot with more detail...

(DSO Overshoot closeup)
The 1.5 second time is closer to actual.

It seems clear that a pristine OPHIR 20C head's output does NOT
produce this Ringing seen in the Rubicon Reviewer's Graphs...


OVERSHOOT

The other issue I noticed is the OPHIR heads output Overshoot...
In the Reviewer's 1.5W graph the overshoot is ~110mV with P-P swings
of about 210mV (or mW).

Blord's Graph shows an overshoot of only ~15mV and a single P-P
swing of ~15mV (or mW).

It should be noted that the higher the Laser Power tested with an
OPHIR 20C head the higher the overshoot will be.


RESOLUTION/JITTER

The last issue is the jittering of the signal over a long period shown
here.

It looks like the jittering is ~1-2mW on the 1.5W graph....

I find it strange that an LPM with a claimed 0.1mW resolution has jitter
that is at least 10 times higher than the resolution.

Perhaps what is causing the ringing is also causing the jitter...


CONCLUSION

From the previously posted data and my own DSO tests it seems
clear that
Blord's graph is much closer to the actual DSO readings...

It is evidently clear to me that there are some major issues with the Rubicon
LPM that need to be addressed and/or fixed before it is ready for the
consumer.

The problem could lie in one or more of the following...

1) the Lasers now exhibit a ringing output since Blord's Graph
1) the OPHIR head is defective or not adjusted properly
2) the Rubicon input electronics is producing the artifacts
3) the Rubicon Firmware is producing the artifacts
4) the Peregrine Software is producing the artifacts



It is important to state that I am only going by what was posted and can only
assume the data is correct since the sellers did not comment on the Graphs
artifacts in the Review thread.

It is also to be noted that at this time I don't have a Rubicon to do any real
time tests or to find out exactly where the problem with the Rubicon lies nor
do I care to help to troubleshoot it's apparent issues in light of the last month's
childish antics by the sellers.

Since I don't have a Rubicon to test (at the moment) I can't verify/test if the
above issues also cause problems with accuracy. So for now we'll assume the
power readings are accurate.



Jerry

 

[Trevor]

I noticed this too and was fairly concerned. I contacted Livinloud to ask him what conditions the graphs were taken under.

He informed me that he held the laser by hand, and that the initial "ringing" you're seeing was due to the beam moving on the surface of the sensor as he centered it. Similarly, the "jitter" you note is also caused by slight movement of the laser on the sensor. These "issues" are caused by factors outside of the LPM and sensor.

( For anyone not aware, moving a laser's beam around on the surface of a radial thermopile sensor like the Ophir 20C will cause fluctuations in the reading. )

It should be noted, however, that Livinloud purchased a Rubicon without an Ophir sensor from us, so I can't speak for how the Ophir sensor is configured.

I'll forward this thread to Livinloud so that he can voice his opinion.

Trevor

 

[ARG]

I find it strange that an LPM with a claimed 0.1mW resolution has jitter
that is at least 10 times higher than the resolution.

While the LPM does have a 0.1mW resolution the Ophir head is rated at +/-0.5mW noise. Just something to keep in mind.

I too contacted livinloud about the graphs when I saw them, as I wanted to make sure that there was no problems with the Rubicon.

This was his reply:

was a fan on as well. my 532nm do the same thing but thats prob due to DPSS module. my 650nm C6 is like a flat light super stable and that was just sitting on the box

He had a fan on, which resulted in unstable readings, and that coupled with the fact that we was holding them by hand is the issue here.
The problem is one that you did not suggest, user error.

 

[Livinloud]

I will confirm this even though alex and trevor have already done it. These were my first tests with my LPM and I had the fan on since its FL and hot as balls in the summer. Didnt realize at the time that something as simple as a fan would cause jitter in the graphs. You should have contacted me to inquire more before posting. Im not trying to be rude at all Jerry since I believe you are a very smart man and can create a great product.

You never contactes me to inquire about the testing conditions while alex and trevir both did. In the end this thread is useless since the jitter was caused by the fan.....actually two fans since I use an overhead and box fan in my bedroom.

Sincerely,
Sean

 

[Trevor]

I hate to play this game, but here I go. :tired:

I did a controlled test, just to put this to bed once and for all.

Here's a video of the test. I recommend viewing it in fullscreen 1080p.

Here's the graph I generated. Click it for full-size.



And lastly, here's the actual dataset for your perusal.

https://dl.dropboxusercontent.com/u/6083542/Rubicon_Data.csv

That should alleviate any concerns you have...

Trevor

 

[Livinloud]

Since my LPM seems to be the question of this conversation I figured Id post the most recent graph produced from it under more controlled conditions (AC was still on since you have to be NUTS to turn that off in FL)

Anyways, graph is perfectly fine

Previous "jitter" was caused by user error. I just saved you $300 Jerry, no need to purchase a LPM and test yourself since you will get the same results. Unless you want to spend $300 and purchase one and then waste your time just to get the same results, choice is yours

Cheers,
Sean

EDIT: just in case someone wants to make the argument, well its not the same laser as the one tested in the OP, well here you go! Blord's 501b tested with my new G1 lens

 

[Livinloud]

And just to put the final nail in the coffin; My most recent test conducted less than 10 minutes ago for kmm5117 of his Sinner MS-SSW 9mm build. Full 1 min test with 100mW of sag in output and look at that, smooth graph when tested in controlled conditions :beer:



Game, set, match. Case closed
Sean

 

[Trevor]

...no followup? :undecided:

Trevor

 

[Livinloud]

Nah none needed when the proof is undeniably good :beer:

 

[Meatball]

Hey.

Just noticed this - but what's with the bump in the end of the power rise?

It looks like overshoot to me. Ringing is just repeated overshoot - but there still seems to be one "ring" where the reading is read too far. The immediate decease in apparent power after that peak doesn't look like a thermally induced power reduction in the laser beam since that's already evident by the long-scale negative slope.

This spike will also appear on systems where predictive figures are summed into the output. Like if you took

(some small constant X derivative of increasing power reading) + current power reading

to get the output on scope.

Any circuitry in there with series capacitance, or heh, even a differentiator could produce this overshoot if the "some small constant" is too large of a value.

On this tiny signals, I wonder if its got to do with metallic head of the ophir. Sure - placing the head on a metal surface, or just touching the head changes the readings understandably - but it also increases the capacitance between the internal circuitry and the head. But that's only possible if the ophir's internal circuitry in supposed to be electrically isolated from the metal casing.

I don't for the life of me know why a fan would produce such periodic ringing only at the start of a signals stability, and not for the long term signal stability. The ringing just doesn't look like thermal or electrical "noise" to me.

Surely air currents can't oscillate the temperature of the thermopile so quickly and so temporarily. And surely electrical or E&M interference from a fan would persist into the steady state of the signal would it not?

What do you guys think?

 

[Trevor]

The overshoot right after the rise is characteristic of all Ophir 20C-A heads.

Trevor

 

[djQUAN]

^yes I also have it on mine. But one thing I noticed is the bump before the output drops (post#6). I'm guessing that was because of the laser being moved when pressing the tailcap to turn off?

 

[Trevor]

^yes I also have it on mine. But one thing I noticed is the bump before the output drops (post#6). I'm guessing that was because of the laser being moved when pressing the tailcap to turn off?

I expect that is the case.

Trevor

 

[Livinloud]

All the bumps at the end are turning the laser off. Bump at beginning is from moving the beam stop and letting the beam hit the sensor. I turn laser on with beam stop blocking sensor and move beam stop to avoid rough starts and jitters.

 

[Trevor]

I think it's pretty disgraceful to just run away from discussion when proven wrong.

It shows the intent of this thread was not scientific discourse... just another attack as a part of Jerry's normal anticompetitive behavior.

Worthy of another red bar, IMO.

Trevor

 

[ARG]

If Jerry was genuinely concerned he would have messaged livinloud before making this thread.

I have a feeling this thread was created to try and cause problems for his competition, and not because he actually cares.

Jerry has a habit of abandoning threads after being proven wrong, I wouldn't expect a reply.