Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

SOLD: LaserBee I w/ Ellipsis $100

Trevor

0
Joined
Jul 17, 2009
Messages
4,386
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

OK thanks for the explanation of the new chart in
Post #*

That does not address the Graph data you posted in
Post #1 that I addressed in Post #7 and Post #13...:thinking:

Jerry

It's just an anomaly, I guess. Here's the two original graphs:

Link - Graph 1
Link - Graph 2

As you can see, the 100% power response is exactly the same when tested under the same conditions. In this case, Ellipsis rendered jitter between 89mW and 90mW as 89.5mW, as it should. The first two graphs can't be compared for response time because the conditions were different - i.e. I was moving the laser at different times and for different durations to show the improvement in the reading of a true power value and smoothness of the data output.

Otherwise, the diode laser used had just warmed up slightly, so it output ever-so-slightly less power.

Trevor
 
Last edited:





Joined
Sep 20, 2008
Messages
17,622
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

It's just an anomaly, I guess. Here's the two original graphs:

Link - Graph 1
Link - Graph 2

As you can see, the 100% power response is exactly the same when tested under the same conditions. In this case, Ellipsis rendered jitter between 89mW and 90mW as 89.5mW, as it should. The first two graphs can't be compared for response time because the conditions were different - i.e. I was moving the laser at different times and for different durations to show the improvement in the reading of a true power value and smoothness of the data output.

Otherwise, the diode laser used had just warmed up slightly, so it output ever-so-slightly less power.

Trevor

Really...... here we go spinning again.
All I know is what I see in the Graphs on you 1st post...

Anything after the fact I would question since you are trying
to sell this weird LaserBee I that you played with repeatedly.

There is no jitter in a non tampered with LaserBee I. It must
have been your Laser or perhaps something you did to the
LaserBee I. Perhaps you over voltaged the Sensor Input
inadvertently with your tinkering and damaged the ADC...

Just because your Peregrine software averages out any true
data does not make a non stable Laser Jitter less. It just skews
the true data readings by placing a band-aid on it to smooth it out.

I would have put all my ducks in order before posting...:whistle:


Jerry
 

Trevor

0
Joined
Jul 17, 2009
Messages
4,386
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

Really...... here we go spinning again.
All I know is what I see in the Graphs on you 1st post...

Nothing is being spun here - I am presenting graphs and data to back up what I'm saying.

Anything after the fact I would question since you are trying
to sell this weird LaserBee I that you played with repeatedly.

There is no jitter in a non tampered with LaserBee I. It must
have been your Laser or perhaps something you did to the
LaserBee I. Perhaps you over voltaged the Sensor Input
inadvertently with your tinkering and damaged the ADC...

My LaserBee is calibrated so that a change of 3.61 ADC counts is a change of 1mW. When a laser's power hovers at a boundary between power values, the arithmetic that determines the reading will sometimes flip to the higher value and sometimes flip to the lower value, as the ADC does the same. That's the nature of an ADC as it hovers right between values - sometimes it goes high, and sometimes it goes low.

But the high and low value are NOT the true value. The true value is derived by sampling the flipping that is occurring back and forth to determine where the true value is - it lies between the high and low values.

It's just like PWM, only in reverse. In fact, that's how a delta-sigma ADC works!

My experiments with the ADC involved using a 12-bit DAC to start at 0V, and gradually ramp up the input, stopping when the LaserBee was close to reaching the top of its range - it was never maxed out. If you check the MCP3221 datasheet (the ADC you used), it notes that no damage will be caused until the analog input reaches Vdd + 0.6V; during my experimentation, the analog input to the ADC reached nowhere near that level.

Just because your Peregrine software averages out any true
data does not make a non stable Laser Jitter less. It just skews
the true data readings by placing a band-aid on it to smooth it out.

Peregrine does not do any averaging reading-by-reading; look at the source code. It's all laid out, transparently.

Are you referring to the "77.7mW" noted on the graph in the OP? If so, that's the average of all the readings that Peregrine logged. I believe EagleEye also calculates a total average of all readings streamed from the LPM.

Ellipsis, on the other hand, acquires additional samples to ensure the returned reading is as close as possible to the true value. Every scientist worth his or her salt does repeated trials.

I would have put all my ducks in order before posting...:whistle:

They are, don't worry!

Trevor
 
Last edited:

ARG

0
Joined
Feb 27, 2011
Messages
6,772
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

I've looked at the Peregrine source, I can also confirm there is no averaging in it.
 

Trevor

0
Joined
Jul 17, 2009
Messages
4,386
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

Okay, Jerry. I ran a few five minute tests to prove my point.

First, here is the profile of the laser from Ellipsis over the course of a five-minute test: http://i.imgur.com/8RRIigl.png

Second, here is the output of the stock firmware over five minutes: http://i.imgur.com/5Ne0pVw.png

(My air conditioner turned on at the end, though that power drop it caused isn't relevant to the issue at hand.)

We'll cover the graph Ellipsis output, first. As you can see, it exactly follows what you would expect of a well-heatsinked diode laser. It powers up, peaks immediately, then slowly drops down to a slightly lower power as the heat equalizes. This is what the graph "should" look like, according to what we know of diode lasers.

The stock firmware does the same, but at a much lower resolution. You can see it cycling back and forth between 90mW and 91mW - because the true reading is hovering between the two, at the peak of the laser's power at startup. Occasionally the ADC will return a value just over 91mW - so it is truncated and sent to the screen and the datalogger. Then, as the value falls a little bit, the LaserBee starts only returning values between 90mW and 91mW - and it begins truncating those down to 90mW. I can't be 100% sure, but I'm pretty certain the values are truncated.

Now, the most compelling piece of evidence, now that we've determined that the ADC is perfectly functional and has no "jitter" caused by damage (just "jitter" from the firmware).

Check this graph out:

Ellipsis / Original Firmware

9QKiZAY.png


As you can see, as the laser warms up, the LaserBee reads it in steps of 1mW. But you'll notice, the original LaserBee firmware steps up under the curve produced by Ellipsis. In fact, the original LaserBee firmware only steps up after Ellipsis has arrived at the same value - showing that the original firmware calculates a value then truncates it.

As the sampled power decreases slower and slower, the curve produced by Ellipsis gets less and less steep, and the steps produced by the original LaserBee firmware get farther and farther apart.

Then, it reaches the peak. Ellipsis reads about 90.8mW through this process - meaning that it is sampling 91mW most of the time, and 90mW some of the time. The original firmware responds to this hovering by oscillating back and forth between the two values.

Once the laser's power begins to decline, it drops low enough so that the original firmware begins truncating down to 90mW, while Ellipsis accurately shows the slow drop in the laser's power.

So, in conclusion:
  • The ADC in this LaserBee is not damaged.
  • Any "jitter" in the output of the original LaserBee firmware is due to the way it treats values hovering at a threshold.
  • Ellipsis produces curves of changing laser power, while the original firmware does not even attempt to.
  • Ellipsis, therefore, produces higher quality data that can, in turn, be interpreted into higher quality information.

Trevor
 
Last edited:
Joined
Sep 20, 2008
Messages
17,622
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

Ellipsis / Original Firmware

9QKiZAY.png



Trevor
That graph just goes to prove that you did indeed damage the ADC
and your "elipsis" firmware hides the fact by removing the ADC jitter
by averaging or by another algorithm Band-Aid.

As I've said before....I've said about all I need to say on this...
no need to continually repeat myself and :horse:

In case it was not clear to some who posted on this Thread.....

Legal Disclaimer

Please be advised that this disclaimer is for Legal reasons and
to be absolutely clear.

J.BAUER Electronics is the sole designer and manufacturer of all
the LaserBee Laser Power Meter products and is the sole owner
of all copyrights and trademarks for the LaserBee™ LPM products
and the EagleEye™ software.

Anyone that implements/uses any unauthorized nor approved nor
tested (by J.BAUER Electronics) 3rd party firmware on any of our
products will unconditionally void any Warranty and/or any present
or future Customer Service for their LaserBee product.

This or any future unauthorized 3rd part mimicking of our Intellectual
Property (IP) and work product is not recognized, recommended
or approved to be used on any LaserBee product ever manufactured
by J.BAUER Electronics.


Jerry Bauer Pres.
J.BAUER Electronics




Jerry
 

Trevor

0
Joined
Jul 17, 2009
Messages
4,386
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

That graph just goes to prove that you did indeed damage the ADC
and your "elipsis" firmware hides the fact by removing the ADC jitter
by averaging or by another algorithm Band-Aid.

That graph says a lot more than that, and you know it.

I'll generate a graph of the raw ADC output to show it's not jittering. I'm growing tired of continually providing evidence to back up what I'm saying, while you are just making unsubstantiated claims about a device that you do not have in your hands.

You have ignored the large body of evidence proving it. But, I am happy to continue generating evidence that proves my point!

Trevor
 
Joined
Sep 20, 2008
Messages
17,622
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

Without having the actual laserbee and firmware in hand there is no way you could possibly know that let alone prove it. You've been very defensive about your products in other threads and in this thread it just appears your bashing. That's the hard knocks of a free market, gotta take it with a smile :)

I think I would know if our similar Brand New un-tampered with
LaseBee I's exhibited jitter or not. They do not...

The graphs of the tampered with LaserBee I shown here obviously
do have jitter with our non-band-aid added Firmware.

The Graph posted by Trevor are quite clear to us...

If you feel that you want that LPM then by all means buy it but we
will not service this particular Laserbee I with a damaged Thermopile
due to the tampering done to it.
We can't stop anyone wanting to buy it. We are just making sure
that the buyer knows exactly what he is paying for and that it is
no longer a supported LaserBee product. :):):):):)

I don't know why this is so difficult to understand....:thinking:



Jerry
 

Trevor

0
Joined
Jul 17, 2009
Messages
4,386
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

I've been presenting evidence and making claims based off of that. Jerry, you are just making unsubstantiated claims. Please provide evidence that contradicts what I am saying. That would be scientific.

I'm going to take this as a teaching opportunity, because, Jerry, I must assume that you are not aware of this particular attribute of analog-to-digital converters. I make this assumption because, if you did know about this, it would mean that you have spent the entire thread lying to prevent any sale of this upgraded LaserBee I. I know you would not stoop to that level, so I must assume ignorance.

--------------------------------------------------

There are a few key claims in play here:

  1. Jerry's claim that my LaserBee I's analog-to-digital converter (ADC) has been damaged.
  2. Jerry's claim that the ADC in the LaserBee I is an ideal ADC, that produces no jitter at a transition point.
  3. My claim that no ADC will have a perfect, sharp transition from one value to the next.
  4. My claim that Ellipsis can use this attribute to increase its resolution.

And here is my source: Analog Devices : Analog Dialogue : ADC Noise

Let's deal with the claim of the ideal ADC, first. An ideal ADC is designed to to return only one value for a given range of inputs. It theoretically translates a linear analog voltage into a perfect stepwise approximation of the input.

However, there is no such thing as an ideal ADC. Take a look at the diagram below, taken from Analog Devices' website:

X5knlaM.jpg


And another diagram, from here:

xnJDm3c.jpg


The diagrams illustrate the concept that, near a transition in an ADC, the returned value can flip to a higher or lower code, with probability proportional to the distance of the analog voltage from the actual transition.

That aspect of an ADC is called Input-Referred Noise, or Code Transition Noise. According to Analog Devices' page, "All analog-to-digital converters (ADCs) have a certain amount of input-referred noise — modeled as a noise source connected in series with the input of a noise-free ADC." Thus, even a so-called "noise-free ADC" measuring a "noise-free" signal would exhibit a level of code transition noise.

Therefore, the LaserBee I ADC cannot possibly be an ideal ADC that produces a perfect stepwise output with no blurry transitions.

Claim 2 debunked.

Now, let's look at a graph of the raw, unprocessed output of the ADC in my LaserBee I. These are ADC values that would ordinarily be translated into power values (for my LaserBee I, an ADC value of 361 would be 100mW).

Q6w0Y42.png


As you can see, it shows the classic behavior of a diode-based laser; an early peak, and a slow decay of power as heat tries to equalize. This dataset was measured at about 77Hz - I wanted to show this blurry code transition as clearly as possible.

Here is a zoomed region of this graph:

sRP8Vic.png


The red arrow labels the peak power of the laser, and the blue arrow indicates the area where the slow drop to the next lower ADC value occurred.

First, let's look at the area labeled in red. You can see that the jumps the ADC makes to the next higher count are sparse, then get denser, then become sparse, then nonexistent. This shows that the actual analog voltage value was approaching the next ADC count (330), but was not yet past it, so it would still return the lower value (329).

Now, let's look at the area labeled in blue. This area shows very distinctly a transition from from a solid value of 329, to sparse occurrences of 328, then a dense area where the split is about 50/50, then 328 gets more and more common until 329 is no longer returned. This is a prime example of the idea shown in the original graphs:

X5knlaM.jpg
xnJDm3c.jpg


The transitions of a non-ideal (that is to say, normal) ADC will be blurry due to code transition noise. The behavior of the ADC in my LaserBee I is a textbook example of this facet of ADC behavior. It does not exhibit any behavior outside the normal operating characteristics of an ADC.

Therefore, the ADC in my LaserBee I is not damaged.

Claim 1 debunked.

Now, having debunked the two key claims Jerry has made, that proves my first claim:

No ADC will have a perfect, sharp transition from one value to the next.

Claim 3 proven.

Lastly, let's look at how Ellipsis can use this attribute of ADC's to increase the effective resolution and squeeze higher performance out of a LaserBee I.

The transition labeled with the blue arrow lasted about 22 seconds:

sRP8Vic.png


Ellipsis samples at about 10Hz, with each actual reading being composed of a large sample of ADC values. By averaging those using the techniques outlined in the paper from Analog Devices (link), that change in reading can be mapped to a much higher resolution than the stock LaserBee firmware.

And that brings us back to this graph, I posted earlier:

Ellipsis / Original Firmware

9QKiZAY.png


The code transition noise present near each ADC step enables Ellipsis to accurately render the performance of the laser as a smooth, 0.1mW resolution line.

Claim 4 proven.

Any questions?

Trevor
 
Last edited:

Trevor

0
Joined
Jul 17, 2009
Messages
4,386
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

Since I seem to have cleared up Jerry's confusion... bump! :D

Trevor
 
Last edited:
Joined
Sep 20, 2008
Messages
17,622
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

Trevor…Trevor... Trevor... you are assuming again. You know what they
say about people that assume...:whistle:
Some of us actually have a full time job. Sorry…I can’t play with you 24-7…. :cryyy:

Just because I don’t post does not mean I agree with you…

Nah… No questions… I already have all the info I need… No I’m not in the least bit
confused. :crackup:

I just have a few statements before I let this Thread die….

You have repeatedly tried to show that our original LaserBee I firmware running on
YOUR tampered with LaserBee I has an excessive amount 1mW jitter as compared to
your mimicked “ellipses” firmware.

My actual Graph and Raw data file of an un-tampered with brand new LaserBee I
clrearly show this 1mW "jitter" claim to be not true.

GRAPH #1
42375d1372766195-re-help-1-lb-i-test.png


attachment.php



BTW... this is what a latest generation LaserBee I looks like.

42374d1372766073-re-help-1-laserbee-i.jpg


Your 1st Post graphs of our firmware and yours show obvious response time differences
with your Firmware taking 50% longer and when questioned about it you post new
graphs showing no response time differences after the fact.
Those 1st post graphs are YOUR “Perigrine” output graphs not ours.
One set of graphs must be some contrived data or an outright lie. But I’m sure you
wouldn’t do that ..:whistle:
So which is it Graphs behind door #1 or the graphs behind door #2 that are the lie.
Sorry stating it was “an anomaly” just won’t cut it….It’s not “scientific” as you put it..:p

Jerry's claim that the ADC in the LaserBee I is an ideal ADC, that produces no jitter
at a transition point.
BTW… don’t even try to put YOUR words into my mouth to try to make yourself look
better and/or try to prove your contrived lie.
I never said that the ADC in a LaserBee I is an ideal ADC, that produces no jitter at a
transition point.
You came up with that nugget of Pseudo-Truth yourself.

You know exactly what I said. In case you are confused I’ll clarify my statement to be
perfectly clear…
I said that an un-tampered with brand new LaserBee I running on approved Data Logging
software like EagleEye exhibits no jitter as YOUR tampered with LaserBee I running on
YOUR “Perigrine” Software seems to show. (see EagleEye Graph #1 above)

All your graphs whether reading our firmware or your mimicked “ellipsis” firmware uses
the same ADC on your tampered with LaserBee I.

There is an outlined procedure by the manufacturer of the ADC in the datasheets on how
to read the numerical data that an ADC can output. That procedure is the same for anyone
wanting to read the ADC data lines in firmware or otherwise.
That ADC numerical data is the SAME for ANYONE for any specific input voltage that is in spec
if the ADC has not been compromised.
You are also bound by those same ADC manufacturers procedures to read the ADC’s data.
That is the TRUE data.
What you do with that True data is anyone’s guess…but it is not the True data as is
evident in your charts.


As to the obvious…..

You state the obvious to anyone that has worked with ADCs or has done the smallest
amount of research. But it is still good info for the members that did not know.

Yes I am quite aware of an ADC’s transitional variations/errors. I became aware of that
~6 years ago (when you were ~16…I believe :p) when we did research on ADC’s and
reading as many data sheets we could before designing the 1st generation LaserBee I.

That is why we chose a 12bit ADC (4096 counts) to read 1000mW instead of using
an on-chip 10bit ADC (1024 counts) to read 5000mW like a Kenometer Pro or USB used.

This requires ~4 ADC counts to change the LaserBee I’s readings by 1 mW effectively
negating the single ADC count variations/errors since it takes a full 4 ADC counts to
make a reading change of only 1mW.. (for your LaserBee to show any 1mW jitter the
ADC count needs to be > or < 4 counts not just one as you elude to.)

As I’ve stated on numerous occasions before…(even though you have denied this)
your own link to Walt Kester’s paper states that it is averaging multiple real time readings
to get a stable averaged reading that you are in fact doing.

Did you also read this part… We did... 6 years ago…
Now, consider the case of an ADC that has extremely low input-referred noise, and the histogram shows a single code no matter how many samples are taken. What will digital averaging do for this ADC? This answer is simple—it will do nothing! No matter how many samples are averaged, the answer will be the same.

If your input signal is clean you don’t need to manipulate the ADC data in Firmware to clean
a non existent or minimally noisy signal. That’s just a band-aid for a lousy hardware design.
I know you needed to do this denied averaging on the Kenometer Pro and USB firmware
and Software for obvious reasons.

The LaserBee I and in fact all LaserBee products are designed to have acceptable clean
low noise input signals.


GRAPH #2
sRP8Vic.png


Looking closer at your Graph#2 of the RAW ADC OUTPUT I see that you are showing
a tremendous amount of 1 count jitter. That is not new to us. Like I said we knew that
6 years ago and did something about it by designing a very low noise input circuit and
using an appropriate OpAmp and ADC to do the job properly.

If your LaserBee I with our Firmware is jittering like shown in Graph #3 then it must be
defective since a new LaserBee I does not have those issues.


In your lengthy post trying to show how knowledgeable you are and how clueless and
a liar I am… you are trying to make us believe that the 1 ADC count variation/error is
the cause of our firmware Jitter on your tampered with LaserBee I… PUULEASE

I still say it is probably a damaged ADC or even the OPAmp or your own “Perigrine”
software that is causing the Jitter in this Graph #3.

GRAPH #3
attachment.php


If it’s not the ADC or OpAmp issues that I mentioned earlier then it is probably your “Perigrine”
Software at the root of YOUR jitter issues.

Since we are unable to replicate your jitter with a new un-tampered with LaserBee I running
on EalgeEye that was designed for our LaserBee LPM products we would believe that
one or more of those multiple problems may be the cause of your Jitter issues.

Note:-
These are some of the issues and the reasons we will not endorse or support any 3rd
party Firmware or Software that has not been tested and/or verified by J.BAUER Electronics.

Not only because 3rd party Firmware may damage our LPM hardware by improper use
of our circuit design but also because untested 3rd party Software may not work correctly
with our LaserBee LPM products as (obviously to us) in this case.


BTW… I don’t want to play your "game" with you any more Trevor…
I have some real world work to do… Don’t forget to have fun…:beer:



Jerry
 

Trevor

0
Joined
Jul 17, 2009
Messages
4,386
Points
113
Re: FS: LaserBee I w/ Ellipsis $150

NOTE: The graphs below are all graphs of unadjusted, raw ADC values to demonstrate the ADC characteristics. To do these tests, I removed the scaling and nonlinearity adjustment in the firmware. These are NOT power graphs, as the graph axes say. I just made Ellipsis spit out ADC values in a format Peregrine could read, to ease the process of data collection.

These are ADC value graphs, NOT power graphs.


My actual Graph and Raw data file of an un-tampered with brand new LaserBee I
clrearly show this 1mW "jitter" claim to be not true.

*snip*

Your 1st Post graphs of our firmware and yours show obvious response time differences
with your Firmware taking 50% longer and when questioned about it you post new
graphs showing no response time differences after the fact.
Those 1st post graphs are YOUR “Perigrine” output graphs not ours.
One set of graphs must be some contrived data or an outright lie. But I’m sure you
wouldn’t do that ..:whistle:
So which is it Graphs behind door #1 or the graphs behind door #2 that are the lie.
Sorry stating it was “an anomaly” just won’t cut it….It’s not “scientific” as you put it..:p

Both sets of graphs were generated using the same laser, on the same LaserBee I. The first set of graphs (the ones you commented on initially when you noted the response time) were created by me repeated moving the laser on and off of the sensor.

The second set of graphs was created by doing an actual controlled test, to be scientific and eliminate all of the variables (such as the moving laser).

The second set of graphs clearly shows the response time is the same.

Plus, the graph you posted and the graph I posted are fundamentally not comparable. Why? Yours is sampled at less than 1Hz (about 0.83Hz, based on the 1.2s interval). The graph of the raw ADC output that I posted is sampled at 77Hz.

To illustrate this, here is a graph of my ADC sampled at 1Hz:

eEM6eX2.png


Now at 10Hz, approximately the LaserBee sampling rate:

bEgFWOq.png


As you can see, the ADC hovers between two values, but does not produce much jitter. As you say, this would not really affect the readings of the original firmware.

None of my graphs are "lies." Variables present in the first set of graphs invalidate them for the use of making a scientific response time comparison.

BTW… don’t even try to put YOUR words into my mouth to try to make yourself look
better and/or try to prove your contrived lie.
I never said that the ADC in a LaserBee I is an ideal ADC, that produces no jitter at a
transition point.
You came up with that nugget of Pseudo-Truth yourself.

Well, that certainly seemed like what you wanted us to believe. Now that it has been established that it is not, great! Now we can move on to more important matters.

What you do with that True data is anyone’s guess…but it is not the True data as is
evident in your charts.

I'll be verifying this LPM against a NIST-traceable LPM and recalibrating it. This will prove how "true" the data Ellipsis outputs may or may not be.

This requires ~4 ADC counts to change the LaserBee I’s readings by 1 mW effectively
negating the single ADC count variations/errors since it takes a full 4 ADC counts to
make a reading change of only 1mW.. (for your LaserBee to show any 1mW jitter the
ADC count needs to be > or < 4 counts not just one as you elude to.)

This statement is a bit disingenuous.

As my graph of Ellipsis vs. the original firmware showed, the original firmware truncates, rather than rounds. So whenever Ellipsis moved up to the next whole number, the LaserBee immediately followed.

Again, I will reference my 10Hz graph:

bEgFWOq.png


As you can see, the ADC hovers between 330 and 331 at the laser's peak power. As you said, it takes about 4 ADC counts to change the power reading returned by the LaserBee - so, every 4 counts (about 25% of the time), a change of 1 ADC count would change the reading displayed by the LPM.

This explains how, even when the ADC is operating normally, you could see a reading bounce between 2 power values.

These are hypothetical values, but the idea is easily illustrated:

300 ADC = 80mW
301 ADC = 80mW
302 ADC = 80mW
303 ADC = 80mW
304 ADC = 81mW

So, if the laser hovers between ADC values of 303 and 304, the reading on the LaserBee's screen would change... with a difference of only one ADC count.

If the original firmware does NOT display that behavior, then there must be some sort of "fix" applied to it to stabilize the reading.

Jerry, to prove your point, why don't you post a laser being tested while the LaserBee outputs the raw ADC values at a frequency higher than 50Hz? I did, and it showed an ADC performing as intended. Why don't you provide a counterpoint?

Lastly, here's a graph of my ADC being sampled at a lower power. As you can see, the power settles on an ADC value of 59, and does not move:

n95QzMn.png


This is because the voltage given to the ADC fell solidly on a value of 59, rather than closer to a value of 60 or 58. Again, this is displaying normal ADC behavior.

As I’ve stated on numerous occasions before…(even though you have denied this)
your own link to Walt Kester’s paper states that it is averaging multiple real time readings
to get a stable averaged reading that you are in fact doing.

Did you also read this part… We did... 6 years ago…

The Kenometers required far more finesse than this.

If your input signal is clean you don’t need to manipulate the ADC data in Firmware to clean
a non existent or minimally noisy signal. That’s just a band-aid for a lousy hardware design.
I know you needed to do this denied averaging on the Kenometer Pro and USB firmware
and Software for obvious reasons.

The LaserBee I and in fact all LaserBee products are designed to have acceptable clean
low noise input signals.

So - your signal is noise-free? Great!
And your ADC is noise-free too? False!

Every ADC exhibits input-referred noise, or code transition. I showed my LaserBee I ADC exhibiting a normal amount of code transition noise (I even sampled at 77Hz to clearly show the effect). This is completely normal as an ADC approaches a transition.

Why don't you show the raw ADC output of a LaserBee of the generation I'm selling?

It will prove one of us is right or one of us is wrong.

Looking closer at your Graph#2 of the RAW ADC OUTPUT I see that you are showing
a tremendous amount of 1 count jitter. That is not new to us.

I'm also sampling at 77Hz. If I sampled at only 1Hz like you did, the "jitter" would probably disappear entirely.

If it’s not the ADC or OpAmp issues that I mentioned earlier then it is probably your “Perigrine”
Software at the root of YOUR jitter issues.

Peregrine just graphs what it's given.

Jerry, you have just made more unsubstantiated claims. You've posted graphs of readings at 1Hz, and compared it to a graph at 10Hz of readings and graphs of raw ADC output sampled at 77Hz. We need to compare apples to apples here.

This LaserBee I does not exhibit "one-count jitter;" it exhibits 1 count of code transition noise near transitions where the ADC value is on the cusp of changing. That is what a normal ADC does. If it had a lot of jitter anywhere but the transitions, that would indicate that it is defective. But it does not.

Your reply proves nothing without a rapid-sampled graph of raw ADC output from an "un-tampered" LaserBee I.

But, as I said, this is all pretty moot anyway; this LPM will be verified against a NIST-traceable unit soon. :)

EDIT: Price changed to $125, or $100 + review.

Trevor
 
Last edited:
Joined
Feb 21, 2013
Messages
5,013
Points
0
Re: PRICE DROP: LaserBee I w/ Ellipsis $100

About to buy this based on principle of a good deal!
 

daguin

0
Joined
Mar 29, 2008
Messages
15,989
Points
113
Re: PRICE DROP: LaserBee I w/ Ellipsis $100

Stay with the science gentlemen

Peace,
dave
 

Trevor

0
Joined
Jul 17, 2009
Messages
4,386
Points
113
Re: PRICE DROP: LaserBee I w/ Ellipsis $100

Sold! :)

Trevor
 




Top