Open Source LPM: What interests you?

[ARG]

I'm going to start working on a new Open Source LPM and I want your ideas on what would make you want to get involved.

Would more features make you build one?
Would a simple design make you want to build one?
Would clear documentation make you want to build one?
If you were to build one would you prefer through hole components?
Would you build one if kits were offered?

Essentially I want to know what would interest you, and how I can get people interested enough to assemble their own LPM, improve upon the designs, start designing their own LPM ect.

Thanks for any feedback!

 

[Speedy78]

I would like one through hole on a large PCB with componants and traces labled. Would be nest to be able to see all that. Could put it in a nice acrylic case. I know everyone wants things smaller and smaller but large components on a large board would be sweet.

 

[Blaster]

In self contained box, with display and power supply 9volt battery up to 4 watts ,two 9v batteries if required , fast response time with a thermopile that takes like 3 seconds to 99% max ..IE point and test no pluging into a notebook. I have data logging but only really use average and peak and record it in a seperate log .

 

[Trevor]

I would like one through hole on a large PCB with componants and traces labled. Would be nest to be able to see all that. Could put it in a nice acrylic case. I know everyone wants things smaller and smaller but large components on a large board would be sweet.

I like large boards too, but someone ARG likes to make small ones... :tsk:

Anyhow. To somewhat expand the scope of the OP, I released OpenLPM a couple years back. It saw pretty limited use because it was, well, limited. Basically, it was a datalogging module, unless you actually built an LPM around it. I designed it with a fairly simple front-end configuration for people who weren't interested in how it worked under the hood, but also provided extensive commenting in the underlying code for those who wished to dive in. As far as I know, no one ever did.

I would really like to get some more people interested in writing LPM code - what would it take (in terms of a developer's kit and supporting documentation) to get more people hacking?

Trevor

 

[NKO29]

Well, i consider myself REALLY good at soldering haha so SMD would be excellent. Obviously the OPHIR head would be my first choice simply because of it's accuracy and speed. In terms of Form Factor, i'd like something as small and portable as possible WITH A SCREEN. Highest readings as possible would also be nice, maybe options for dual powering for different ranges eg:
9v battery for 0-4W and an external PSU for 0-20W. I don't care for any USB activity, as i don't see a real use for datalogging TBH.

 

[Blaster]

We think very alike Josh. Point and play ,display, high power capacity and fast response time. Data logging so so ....



QUOTE=NKO29;1246410]Well, i consider myself REALLY good at soldering haha so SMD would be excellent. Obviously the OPHIR head would be my first choice simply because of it's accuracy and speed. In terms of Form Factor, i'd like something as small and portable as possible WITH A SCREEN. Highest readings as possible would also be nice, maybe options for dual powering for different ranges eg:
9v battery for 0-4W and an external PSU for 0-20W. I don't care for any USB activity, as i don't see a real use for datalogging TBH.[/QUOTE]

 

[The Lightning Stalker]

I prefer through-hole because it's much easier to rework if anything goes wrong. I also have a large stock
of through-hole components and they are usually well marked, unlike SMT. Also cost is huge. There is no way I can afford an Ophir head.

 

[Things]

SMD is good fun, once you start you can't go back. So much cheaper than through hole too

 

[The Lightning Stalker]

Oh I know. I have a lot of experience with SM, but NOS through hole components can be had
for almost nothing. Sometimes there are 2 or 3 LM358s in an old computer power supply.
It's unbelievable what you can find in the garbage sometimes, really. The downside is that
the boards have to be drilled, but if it's a one-off build with low parts count, protoboard is a
viable option and it's predrilled.

 

[djQUAN]

Actually, with mass produced junk these days, you can pretty much get a lot of SMD stuff too. With the right soldering iron and a chea hot air rework tool, you can also get a lot of SMD parts for free.

I like to work with both. I like SMD since you don't have to drill boards, bend leads, solder then trim off the excess (and the occasional flying offcuts that go everywhere). But with SMD, you need plated through holes to reach the other layer of the PCB, and you can't do point to point. That's why I have a stock of both.

 

[Meatball]

Plus there are some noise related advantages to using smd resistors over carbon comps.

As long as the parts for the design are well specified, linked to, highly available and very accessible, something like a kit would not even be necessary.

Also, might be neat to use uC sockets instead, so one can easily swap out for a differently programmed controller.

 

[pschlosser]

In my opinion, LPM operation need not be complicated. You point your laser at the sensor and you read the value. What’s so complicated about that? But as I began using them, and measuring various things, I’ve learned I need more than one, because a single LPM won’t do what I need it to do.

Also, an LPM is both a meter AND a measure head. Not just the meter. It’s incomplete without all you need to take a reading.

My LPM needs are met. But if I were starting over, this is what I would want from an LPM:

• Measure Low. I want to measure low-power HeNe’s I buy off eBay. It needs to measure down to 1mW +/- 0.10mW.
  
• Measure High. I want to measure high-power diodes, such as those Nichia 445nm in the >2000mW range.
  
• Simple Setup. It can't involve delicate wires, four to seven components that must be hooked up in sequence in order to work. Simply turn it on, point and read.
  
• Battery Powered. I want to take it with me when I visit other lasers and measure before buying.
  
• Externally Powered. From a wall wart or USB interface, so I can leave it on for long periods.
  
• Peek and Hold. I want to see the maximum my laser diode puts out, before it begins to heat, drain the batteries and the output falls off.
  
• Average output. Between two time points.
  
• Fast read-time. I don’t want to wait 10 or 20 seconds before the reading can be taken. For laser diodes, this is a problem. Not so much for continuous wave.
  
• Graphs. Data collection is fine. But I want to share what my lasers can do with others on the LPF forum. Those nice graph images people post are really slick. I want software to do that.
  
• Cross-platform. I love windows because I can tweak it and I’m a software engineer. But macs are cool, too. And as long as I am far-reaching, why not through an app from a mobile device?
  
• DIY meter head. Sure, Ophir heads are pretty awesome. But so are their price tags. Can we make our own meter heads using inexpensive products? How can we standardize calibration, so we can do it ourselves? There must be a way to calibrate and get it close. And how close is close? Is +/- 1mW good enough? +/- 10mW? Can we get it that close?
  
• Ophir and other heads. When I finally do buy an Ophir head, can I use it with this Open Source LPM?

Pretty crazy I might want all this, huh? But there’s more. I don’t want to be required to install large funky libraries, or .NET 9.0 on my machines to use the LPM and graph output. Further, I want the source code, so I can fiddle with it in my evil lair and improve it in my laboratory. Then when I can graph phase-shifts of my beams off the moon, I can contribute to the Open Source LPM project and become infamous.

In the end, though, the meter-part of this project, and the “Open Source” software involved, need not be all that complex. And many of the above-mentioned features are dependent on the hardware, not software.

What the community needs is an LPM that entry-level DIYers can afford to build, acquire and use. Software features are cool, but it’s the cost of the hardware throwing up roadblocks. In many cases, the cost of an LPM is several, if not ten (10) times more expensive than the product measured. As such, it’s not among the first tools acquired by the hobbyist. But man, does it teach so much. Imagine the advances we might make, if such tools were as freely available as the laser sources themselves.

 

[benmwv]

I think it should be smd, but have the ICs in through hole packages. That way they can be easily replaced. If you use like 1206 for the resistors, caps, etc. it will be easy to solder. I think 1206 is even easier than through hole because you dont have to do all that bending and cutting. Just set it and solder.

It needs to be cheap to get a lot of people to build. And preferably I wouldnt need to buy anything to program it, or you would sell the ICs pre-programmed.

Data logging is a must, idk what you guys are thinking! How else would you get a power graph, by hand?

And finally, self calibration. Write a tutorial with links and whatnot say that if you use this specific TEC, this exact paint, this number of even coats, everything followed exactly as shown, and then you put exactly X volts through this specific resistor while it is in good contact with the center of the TEC you can calibrate it to X mw with approximately X% accuracy. If you can get a +/- 5% or even 10% it would be a useable stand in until the user can get it more accurately calibrated.

It should probably accept 1v/watt (but still keep that easy to change in software), and have an opamp sub board. That way the standard ophir heads are natively compatible and TEC or other are able to be used with the additional amplifier.

 

[ARG]

Everything you guys have mentioned is doable.

Some want data logging, some want just a screen so it will have to have both to please everyone

Also, an LPM is both a meter AND a measure head. Not just the meter. It’s incomplete without all you need to take a reading.

Sure, Ophir heads are pretty awesome. But so are their price tags. Can we make our own meter heads using inexpensive products? How can we standardize calibration, so we can do it ourselves? There must be a way to calibrate and get it close. And how close is close? Is +/- 1mW good enough? +/- 10mW? Can we get it that close?

For now this will just be the head. Ben has been working on a fast response thermopile, but it needs lots of testing before it can be put to use.

The biggest setback DIY LPM's have at the moment is the heads. There definitely needs to be lots of work done there.

• Measure Low. I want to measure low-power HeNe’s I buy off eBay. It needs to measure down to 1mW +/- 0.10mW.
  
• Measure High. I want to measure high-power diodes, such as those Nichia 445nm in the >2000mW range.
  
• Simple Setup. It can't involve delicate wires, four to seven components that must be hooked up in sequence in order to work. Simply turn it on, point and read.
  
• Battery Powered. I want to take it with me when I visit other lasers and measure before buying.
  
• Externally Powered. From a wall wart or USB interface, so I can leave it on for long periods.
  
• Peek and Hold. I want to see the maximum my laser diode puts out, before it begins to heat, drain the batteries and the output falls off.
  
• Average output. Between two time points.
  
• Fast read-time. I don’t want to wait 10 or 20 seconds before the reading can be taken. For laser diodes, this is a problem. Not so much for continuous wave.
  
• Graphs. Data collection is fine. But I want to share what my lasers can do with others on the LPF forum. Those nice graph images people post are really slick. I want software to do that.
  
• Cross-platform. I love windows because I can tweak it and I’m a software engineer. But macs are cool, too. And as long as I am far-reaching, why not through an app from a mobile device?
  
• Ophir and other heads. When I finally do buy an Ophir head, can I use it with this Open Source LPM?

That is a lot of feedback, thanks! I'll keep this in mind.

 

[NKO29]

NOBODY has mentioned this... PRICE. Obviously the overall price of the board would be awesome if it was as low as possible

 

[ARG]

I should be able to keep it around $30 of parts. Is that more than you guys think is reasonable? I can remove a few components and use cheaper parts to get the cost down if need be, but features and quality will be sacrificed then.