DS00 Time Of Flight Laser Range Finder prototype evaluation
prof_braino
Posts: 4,313
A new device has been developed as a result of this thread
http://forums.parallax.com/showthread.php?133632-Time-of-Flight-Laser-Project&p=1077074&viewfull=1#post1077074
Thanks to all the folks that contributed, especially the folks that provided exidence indicating this was (is) impossible.
This once again proves that at least on these forms, the best way to get something done is to declare it impossible, and wait for a couple months.
Incidentally, the assessment that this device is not practical with current technology and we would have to wait about 20 years, was only off by 19 years and 4 months.
So, now I have the prototype in my hands. Time to play!
Here is a link to the draft of the user manual, from the designer
http://www.lightware.co.za/download/doc/DS00%20-%20Laser%20Range%20Finder%20Manual_Rev_00.pdf
There is a data sheet for the DS00VQ100 timing chip, I have to dig backand find that.
EDIT - here is the post that contained the original draft datasheet
http://forums.parallax.com/showthread.php?133632-Time-of-Flight-Laser-Project&p=1027009&viewfull=1#post1027009
A new version has not been issued as of 2012-03-10 (March 10).
The provided firmware is in C (initially I said spin, sorry). I tend to use another language, but I want to thoroughly examine the unit as shipped before I start introducing more variables.
Please feel free to comment and correct the coding errors I'm bound to make as this thread progresses.
PLEASE POST if you have any application that might benefit from a long range high precision laser range finder. This will help influence the direction of the investigations.
http://forums.parallax.com/showthread.php?133632-Time-of-Flight-Laser-Project&p=1077074&viewfull=1#post1077074
Thanks to all the folks that contributed, especially the folks that provided exidence indicating this was (is) impossible.
This once again proves that at least on these forms, the best way to get something done is to declare it impossible, and wait for a couple months.
Incidentally, the assessment that this device is not practical with current technology and we would have to wait about 20 years, was only off by 19 years and 4 months.
So, now I have the prototype in my hands. Time to play!
Here is a link to the draft of the user manual, from the designer
http://www.lightware.co.za/download/doc/DS00%20-%20Laser%20Range%20Finder%20Manual_Rev_00.pdf
There is a data sheet for the DS00VQ100 timing chip, I have to dig backand find that.
EDIT - here is the post that contained the original draft datasheet
http://forums.parallax.com/showthread.php?133632-Time-of-Flight-Laser-Project&p=1027009&viewfull=1#post1027009
A new version has not been issued as of 2012-03-10 (March 10).
The provided firmware is in C (initially I said spin, sorry). I tend to use another language, but I want to thoroughly examine the unit as shipped before I start introducing more variables.
Please feel free to comment and correct the coding errors I'm bound to make as this thread progresses.
PLEASE POST if you have any application that might benefit from a long range high precision laser range finder. This will help influence the direction of the investigations.
Comments
DS00 Unit in carrying case, as shipped. Pelican case signals caution is advised.
Contents of the Pelican Case, as shipped.
* USB cable connects to PC USB for virtual serial connection and power, as the Propeller QuickStart board.
* A connector for adding a battery pack for remote operation
* 2gig USD drive with Manual, Quickstart Guide, Light ware contact info. This will later contain the spin source code for the demo program, and possibly example application programs. Also, the DS00VQ100 datasheet should be included.
* Assorted mounting standoffs and screws, and an Allen wrench
Business end of the Optics module. Dark looking lens is the IR filter on the detector.
Laser board and detector board on Optics Module.
Propeller Board. Looks like there's a 10.000 crystal and 10 LEDs for status. The remiander looks like standard prop chip, EEPROM, FTDI, mini USB
DS00VQ100 chip carrier board. There are six programmable test points and a ground, all with loops for connecting oscilloscope probes.
C source code archive attached
The Pelican case is the first sign how serious a tool this is.
Is it safe? What are the safety concerns and precautions that need to be take for the laser? Usually, lasers consumers encounter are limited to 3 milli watts. This one is about 4,000 times more powerful. This device demands significantly more respect.
The original spec called out Osram, 14W SPL LL85 Laser Diode at 14 watts, 850 nm (which is Near-Infrared, between 780-1400 nm). The prototype is 11 watts. The the prototype and the final unit are characterized as Class 1M *. What does this mean?
Here's link to the OSRAM safety document.
http://catalog.osram-os.com/catalogue/catalogue.do;jsessionid=10B966B06BA0022F9A0DAEB7DB17010A?act=downloadFile&favOid=0200000400010185000100b6
Most of the OSRAM doc is beyond me.
Here's material from other sources.
* From wiki - Class 1M laser means it has a wide or divergent beam, and the Maximum Permissible Exposure (MPE**) cannot normally be exceeded, (as long as its not refocused through for example a microscope or binoculars).
** From wiki - MPE (maximum permissible exposure) is the highest power or energy density (in W/cm2 or J/cm2) of a light source that is considered safe, i.e. that has a negligible probability for creating damage. It is usually about 10% of the dose that has a 50% chance of creating damage
under worst-case conditions.
Class 1M
How can a 14 watt or even an 11 watt laser not exceed the MPE? Some calculations...
The DS00 materials state that the laser is pulsed on for 30 nanoseconds, and this happens at 25kHz. 30 nanoseconds * 25,000 pulses per second is 750,000 nanoseconds on-time = 0.00075 seconds on-time each second. So the laser is off most of the time. Hmmmm. I don't know if I got this right, but it looks to me that 0.00075 second on-time * 11 watts is the equivalent of about 0.00825 watt (8.25 milli-watt) continuous laser. Did I calculate this correctly? Or do we still have to account for RMS power, and multiply by 0.7? If so, I get 0.00075 (ON/s) * 11W * 0.7 = 5.775 milli-watts equivalent continuous power. In terms of power, this is still higher than the 3mW limit for class 1, and the 5mW limit for Class 3R. The divergent beam is what makes this Class 1M. Were it not for the diffused beam, this would be at least Class 3B and would require protective eye-wear, a key switch and a safety interlock. So, I'm still a little iffy about who should be play with this, and how it is to be used.
To summarize the rating, the laser on its own at 11 watts would be Class 4, or totally dangerous. When the laser is powered by the DS00, it is pulsed, not continuous, so the effective power is reduced. Since the optics diffuse the beam, the laser IN THE DS00 UNIT is Class 1M. So its got powerful parts, but the dangers are mitigated by the way they are arranged.
To test, I scrounged an ancient Logitech QuickCam Express. Using an old TV remote, the camera can detect IR as expected. After the QuickCam IR filter was removed, the IR remote was able wash out much of the image on the quick cam. After various preliminary tests, the DS00 optics were pointed directly at the QuickCam at distance of 1cm. This is more than ten times closer than anyone is ever going to get to this thing, so if this doesn't cause damage, we might be ok. After several seconds, the auto white balance stabilized, and the camera stared directly in to the laser. The laser was left on for over 1000 seconds (this number was in the old Class II standard, I just let it run while I did chores).
Afterwards, the QuickCam was able to record regular images and exhibited no damaged.
I'm still not going to let anyone look into the laser. Rule of thumb is going to be (a) do not mess with the optics, and (b) don't point it anyone's face.
From the above, I'm getting that the DS00 unit as shipped is reasonably safe as any other laser device. So long as we do not alter the optics, or look at the device through telescope of microscope, we are OK. [At some distance, the power of any laser should be reduced to the point is is no longer harmful (from inverse square), but I don't know how that works with lasers and optics. Anybody clarify this?]
BUT this is an experimenter's unit. There is a possibility that somebody would decide to modify the unit for a particular application. In that case, all safety concerns are in the hands of the user. Without divergent optics, the unit is Class 4, the most dangerous class of laser. Also, a user may alter the pulse duration. I don't know if this is possible via software or whether is is hard-coded in the DS00VQ100 chip, maybe this will be in the datasheet.
My final feeling is that the DS00 should be treated with similar respect as power tools, like a chainsaw or a welder. This is no toy, but its safe enough for a reasonable person to use.
Any care to comment? Corrections and additions are requested.
I plugged in the USB cable. The DS00 beeped 5 times and the laser fired (detected by the web-cam).
When I unplugged the USB cable (to power down the laser), the PC shut down.
This only happened twice; now that I've done this several more times during the burn-in tests, it has not happened again.
Perhaps this was due to the webcam or FTDI drivers not being fully installed?
A very nice explanation Prof.
The rules for laser safety apply to the "final assembled product" which is why a laser component can be Class 4 whilst the finished product can be Class 1M. Many laser devices, even 3mW laser pointers, now use Class 4 lasers inside but they are down-rated electronically and optically by design. However, you are correct in suggesting that caution is required if people are going to dismantle the optics. The laser optics create a beam that is spread out over 490 square millimeters and then begins to diverge. This reduces the energy density quite considerably, even close up to the optical face. If you change the size of these optics then you will change the safety margins in a counter-intuitive way, with a smaller lens becoming more dangerous and larger one becoming safer.
There is also the issue of using a telescope or binoculars to look directly into the beam from some distance away. The standards do not specify a safe viewing distance because, in theory, there is no limit to the collecting aperture of the telescope used, and therefore, it could collect all the laser energy and focus it onto your eye. The rule remains, don't look directly at the beam with any form of optical instrument - use a camera or a webcam.
The good news is that the laser pulse width and firing frequency are immutably controlled by the DS00VQ100 chip so you cannot accidentally increase the average laser power. Also, if taken out of the optical module, the laser component will have a very high beam divergence, meaning that you would probably need to hold it closer than 2 inches from your eye before you exceeded the 1M power density rating.
I've tested nine different units on two different machines (PC and Mac) with more than 300 insertions and not seen this particular behavior. I will do some more tests on different PCs with different operating systems.
-Phil
I wasn't one who said it was impossible but I sure thought it might be. This is super amazing.
Thanks for the posts Prof and Laser.
Like Phil, I'm curious what the price for this will be.
Did I mention this is amazing?
I have had quite a few requests for pricing so I'm setting up a small website at lightware.co.za with some basic information. This should help those early adopters who want to be first with this new product. In the long run, there might be manufacture and distribution in the USA with a range of models and laser range finder components available, hopefully at lower prices. There are a lot of "if"s between now and then!
Installed minicom
sudo yum -y install minicom
set the baud to 57600
and turned off all flow control.
set port to /dev/ttyUSB0
Plug in DS00, which beeps five times and begins displaying distance measurements at about two readings per second.
The demo firmware is configured properly, so it starts displaying measurements as soon as it connects.
DS00 is totally awesome. Its measuring continuously from 0.17 meters to 19.75 meters.
Within 2 seconds on a target, the readings stabilize. The reading are in meters with two decimal places. The readings vary by 0.01 meter every ten or twenty readings.
I see I had averaging on. With averaging off, it looks like the reading vary +/- 0.03 meters.
The ceiling above is measured as 5.45 meters, with occaisonal reading as high as 5.47 and as low as 5.43. Once in a while 5.83 and 5.36 have also popped up.
The only thing it can't measure is fluorescent light fixtures with 3 tubes of 48 inches. No surprise there. I haven't taken it outside yet, since its raining.
The only things that I would change is perhaps I would want a visible laser while I'm getting my feet wet. While this would reduce the range and effectiveness of the device,I would get a better feel for what is going on, and it would be suitable for my tiny lab. I asked the boss if I could get a new lab that's greater than 375 meters in at least one direction, but I did not get an answer yet.
To see the florescent tubes you can:
1. Manually increase the sensitivity of the measurement by cranking up the APD bias voltage - press V then enter 170 to raise the voltage to 170V. Set it back to 140 when you've finished.
2. You can also use a higher amplifier gain but the calibration will go out by a few inches. Type G and try 14 or 15. Always set this back to 8 when you've finished.
3. You can safely use both of these methods at the same time.
4. The LED display indicates the signal strength. If it's above the bottom then the signal is valid.
To measure over a long range you can:
1. Increase the measuring range limit to 96m by pressing M and entering a setting of 2.
2. To keep the signal strong enough you can boost the gain of the APD and the amplifiers using suggestions 1 and 2 above. Using them both at once will give the best range.
3. At longer ranges the update rate slows down so you can speed it up again by changing the time expansion - type X and enter 250. Put it back to 152 when you've finished.
4. By default the averaging function is on so if you want a quicker response whilst trying to hit a tree 50m away then you can toggle it off by typing A. Switch it back on when you've finished for more stable results.
Have fun!
-Phil
This usually occurs for two reasons:
1. The drivers are not correctly installed, or are in conflict with other drivers.
2. There's a dead short in the USB cable when you unplug it, and the USB hub in your PC just happens to not have a poly fuse on it, or some other cost-cutting measure, and the PSU is shutting down. On my Sony Vaio, if a USB device draws too much current a little bubble appears on the taskbar that says (to the effect of), "Whoa, dude, that USB device just caused a power surge."
-- Gordon
I turned off averaging to measure through window screen.
Building and evergreens were easy to get a reading.
Trees with no leaves were a challenge. I could not really tell if I was aiming properly, maybe I should add a green laser pointer.
So far, the DS00 is pretty darn effective without adjust any settings.
Also, the DS00 is is showing how great the OLPC XO screen is. The XO is perfectly readable in direct sunlight. I had the DS00 measuring distance and the XO displaying the reading for about under an hour. The XO battery is still about 50%. I think the XO is working out well as a field terminal.
Please send me a Private Message (click on my user name) with your contact information direct email, etc and I will get the process started.
http://www.nasa.gov/centers/dryden/Features/ER-2_mabel_mission.html
Is the difference between the 50m kit and the 100m kit the laser, with the former using 11watt and the latter using the 14 watt?
To me, "Kit" says "bag of parts that we must assemble". I expected these to be an assembled and tested functional units. Could this be clarified?
Would "DS00 14watt 100m Range Finder" and "DS00 11 watt 50m Range finder" be better names for these?
Is there going to be any "optics" entries? Right now I don't see any laser, lens, filters, or mountings; is this intentional?
The only difference between the 50m unit and the 100m unit is the optics. The 100m unit uses a 50mm lens to capture light for the detector whilst the 50m unit uses a 25mm lens. There is an interesting theoretical relationship between range and receiver lens size - the range is linearly proportional to the receiver lens diameter. This comes about because the return energy falls off as an inverse square law with distance whilst lens area increases as a square law with lens diameter. These two effects cancel each other out so that you end up with range being directly proportional to lens diameter.
Using the 50mm diameter receiver lens and 100m range as our starting point, we can see that a 25mm lens will give 50m range, a 10mm lens will give 20m range and so on, ceteris paribus (all other things being equal).
A good point. The name comes about because I see the DS00 as a modular bunch of parts that can be modified rather than as a "finished product" which implies a closed unit with a housing that has a "warranty voided if you open it" sticker on it. I'll think about it further - more suggestions welcome.
Yes, well, we're still trying to figure out how to use Joomla!
So using a 187.5mm lens, could the DS00 measure out to the full 375 meters? Could an acrylic Fresnel lens work?
I don't want to compromise the original unit, could a larger lens be placed in front of the stock lens, or would the stock lens need to be removed.
If I used the 1 meter satellite dish that I re-purposed as a telescope reflector using aluminum foil (let it suffice to say that my wife hated that project) could the device measure 2 kilometers? Or would that exceed that device's counters? Would it give improved readings at shorter range? I have a feeling that this is going to scare the neighbors again.
In theory yes.
It might, but...
...there's going to be a point at which this is not the best way of improving performance. Like all engineering problems, if you continue to improve one specification, in this case the aperture of the receiver lens, you will start to run into diminishing returns. For an optical design it is as important to consider the quality of the optical system as it is the size.
To get really good range performance out of the DS00 you would have to remove the existing acrylic lenses and replace them with:
1. a system for improving the beam pattern and divergence of the outgoing laser light (better collimation)
2. improved receiver lenses that have low spherical aberration and anti-reflection coatings (a small circle of least confusion)
3. a series of lenses for directing the light perpendicular to the optical filter (or remove the filter completely and only work in the dark)
With these relatively trivial modifications
And therein lies the proverbial rub. The better that you make the long range performance of the optical system, the worse the short range performance becomes!!!!:frown: This is because a "good", long range lens system usually has worse parallax and shallower depth of field than a "bad" system. So for a good lens system, as the target surface gets closer, the tiny spot of returning light on the tiny optical detector wanders off the detector and starts to get blurry. The signal strength drops completely and the short range performance sucks.
Designing a lens system that works well at both long range and short range is much harder than designing one that works only at long range.
Actually, a bigger lens in front of the laser can be much safer than a small one because the energy is spread over a bigger area. It just looks more scary!
It is available for sale now!
Analog1
That's pretty cool! Did you get get one? I'd be interested to see how it tests out. Does LIDAR mean it can detect the targets relative speed?
The page says it has a range of up to 4 feet, can detect a white wall at 5 feet. And its $400.
Let me know if you have more questions. You can also use the "Contact Info" at the bottom of the web page and get direct support.
I believe you can return the unit within 30 days if you are not satisfied with it.
Analog1
Loading a new program will necessarily overwrite the existing firmware.
So, before start doing experiments, I want to make sure I can return the DS00 to it original as-shipped state.
The imageCraft ICC compiler was use to compile the lightware C source code. It is a30 day trial software
I try to avoid trial software, so I thought I would use the "regular" prop C compiles, the two I hear of the most are Catalina and GCC
Both of these are different from ImageCraft, but a compiler is a compiler and C is C, how hard could it be?
I knew going in I would have to make some changes to the source code, if nothing else to match different name in libraries. This is no biggie.
What I find however, is that I cannot get either one to compile the same way the regular users do.
They can get it to at least build, I cannot.
Here are the threads for both
http://forums.parallax.com/showthread.php?139568-GCC-for-DS00-C-code
http://forums.parallax.com/showthread.php?139573-Catalina-for-DS00
If I use the exact same code, and the exact same compilers, how can I get different results?
Maybe there is something that both Catalina and GCC expect to find on the machine that I do not have installed?
Is ther something we pre-install before we install a C compiler these days?
Perhaps I should start with the imagecraft compiler after all, and try the others after I prove I can make the trial version work at least once.
There has been a pretty interesting response to the DS00 prototype, mostly from folk wanting to buy them. Also, comments have indicated some changes might be made. These are in the works, nothing definite at this time though.
The complication is bringing in a laser from outside the US, we have to do a special ritual to get permission. I'm learning about this, and s-l-o-w-l-y getting the pieces in place.
This is still on-going. All the folks that contacted me, please forgive the slow progress. The laws are picky, and there cam be serious consequences if everything is not "just so". First priority is keeping things legal and safe.
More updates as information becomes available....