Runtime electronics

[NewBie]

Rather low cost 12 bit, 11 channel voltage measurement system to serial port, plugs right into a computer.

64 dollars.

Additionally it has 3 digital inputs and 3 digital outputs, which can be used to turn things off and on (like turn a flashlight on and off, with minimal additional electronics.)

Comes with a data logging utility to import data into other programs and spreadsheets (such as Excel). Quick BASIC demo programs included.

You can use it to monitor input and output, current and voltage, hook it to a light measurement system, and have it shoot it all right into Excel where you can have Excel compute and then graph runtime, power in and power out for efficiency measurements over battery voltage, light output, current to the luxeon, and do it all in one test. Pretty slick setup.

http://www.bb-elec.com/product.asp?sku=232SDA12&PathId=40&altsku=daqaichart

 

[MrAl]

Hello again,

Looks interesting, and 25 samples per second isnt bad
for monitoring battery equipment. My meter is slower
then that yet i got more then enough samples to
calculate runtime, discharge curve, etc.
I think i took one reading per second when i was
monitoring charge/discharge cycles of some batteries.

Take care,
Al

 

[NewBie]

Thanks.

Have you ever set up Excel to directly import data off the RS-232 port, real time? I've seen it done, but didn't have time to check out how it is done.

 

[MrAl]

Hi Newbie,

Well, i dont have Excel so not really :-)

What do they charge for that, do you know?

Take care,
Al

 

[Doug Owen]

Even though it costs nearly twice the money, I own and can recommend Labjack.

It only has 8 channels, but they can be used differentially (a big advange in lots of cases) in pairs. Each channel has a *programable* attenuator so the full scale range can be set by software with Gains of 1, 2, 4, 5, 8, 10, 16, or 20 V/V. It also has two analog outputs, and it's *20* digital lines can be inputs or outputs. Perhaps of most interest to serious types, it can be run under LabView. USB interface (RS-232 is a dead/dying standard, not standard on newer computers).

The provided software outputs it's data in excell readable format, although you have to drag them in or set the program to store them as .xls files.

Doug Owen

 

[MrAl]

Hi Doug,

Wow, that's pretty nice!
I could see making a battery charger for the various
types of batteries with that thing and a simple
external controllable power supply (LM317 perhaps).
Since they provide a software api, you could program
it for just about any charge character you like.
Sure would be interesting...
Im gonna have to think about this one :-)

I've also been kickin around the idea of the little
controller ic's to use for battery chargers too.
Nice thing about them is that they'll make a stand-alone
charger when you're done. Bad thing is they arent
as accurate as that device.

Can i ask what you have used it for in the past
and what the results were?


Take care,
Al

 

[Doug Owen]

I've mostly used it for data logging in long term temperature and current tracking. The spreadsheet output works just fine for what I need. Differential input against a reference supply gives you a bunch of resolution when you really need it (or like me, just want to see it happen.....).

I've given some passing thought about building a run time rig to log battery voltage, current and light output. Maybe even temperature?

All in all a neat thing, but since I'm not a softhead, most of the really neat stuff you could do with it are not available..... Hey, Al, you get one and write the software and I'll check it out. Or don't get one, just write the software and I'll tell you how it works......

BTW, those surface mount regulators came the other day, I gotta dig out the 'surf boards', find the transistors, and build up your circuit. Shouldn't take too long. I hope.

Anyway, LabJack is a solid product IMO. Easy to use terminal strips for all the analog and some of the digital (the rest are on a d sub). Recommended to those it interests.

Doug Owen

 

[MrAl]

Hello again Doug,

Oh yes, the diffy inputs sounded great. Yeah, i bet
with a reference you could get some really great
resolution. Sounds wayyy too cool he he :-)

I guess i should look at the api SDK if i can find it,
i guess it's on the site with the manual? i should have
looked :-)

OH yeah, the surface mount reg's. Will be nice to see
how well they work out.

I found it a very interesting piece of equipment for
the price. The only thing i might question is why the
analog output max rate is about 50 Hz or so, any ideas?
I can live with the 10 bit reso, that's ok i guess.

So i guess what you're saying is that if you had to measure
a battery voltage of 1.5 volts and you had to have
resolution like 1.500, 1.501, 1.502, ..., 1.520
you could use a 1.5 volt reference with the diff inputs
and then what range would you get that way (best)?
Like, lets say 1.5 to what...1.6, or more like
1.5 to 2.5 volts? (which isnt bad either).
I guess what im trying to find out is what would be
the best resolution you could get using a 1.5 volt
reference or even zero volts to what, 1 volt?
0 volts to 1 volt across 12 bits would be something
like 0.0002 volts i think? is this possible?


Take care,
Al

 

[Doug Owen]

[ QUOTE ]
MrAl said:

I found it a very interesting piece of equipment for
the price. The only thing i might question is why the
analog output max rate is about 50 Hz or so, any ideas?
I can live with the 10 bit reso, that's ok i guess.

So i guess what you're saying is that if you had to measure
a battery voltage of 1.5 volts and you had to have
resolution like 1.500, 1.501, 1.502, ..., 1.520
you could use a 1.5 volt reference with the diff inputs
and then what range would you get that way (best)?
Like, lets say 1.5 to what...1.6, or more like
1.5 to 2.5 volts? (which isnt bad either).
I guess what im trying to find out is what would be
the best resolution you could get using a 1.5 volt
reference or even zero volts to what, 1 volt?
0 volts to 1 volt across 12 bits would be something
like 0.0002 volts i think? is this possible?


Take care,
Al

[/ QUOTE ]

First off, I suspect the D/A is kind of an afterthought. It's no doubt coming from the digital IO words, so the relatively low transfer rate makes it a poor solution to waveform generation for instance. However, it may be just fine for plotting V/I curves. You have the option of reading the output on a 'spare' channel and making it a pretty good data point.

On the floating diff inputs front they have a ten Volt (plus and minus) dynamic range, otherwise you need an external divider or other level shifter. Once there, you can sum them (or difference them) and use the PGA function to decrease the working range to match the A/D more closely. The A/D is still 10 Volts full scale. But at a gain of 20 V/V that's only half a Volt FS over the 12 bit range 4 counts per mV. Yup, yer math's good as ever, that's what I recall and that's how I read the specs.

We were trying to track room temps to .01 C, and magnet currents into the PPM ranges. The latter took some tinkering, the former was quite straightforward IIRC.

Nothing real fancy, the sort of resolution you'd expect from a typical 3.5 digit DMM or a bit more. Plenty for most any reasonable task.

Doug Owen

 

[NewBie]

Or you can construct your own.

Analog Devices and TI make some decent A/D parts.

Or you could use a microcontroller with a built in A/D.

Atmel has some USB interface code you can drop right into your compiler. Either that or just hook it to your RS-232.

 

[MrAl]

Hello again,

Thanks Doug, for the info on this thing. It certainly
does sound very interesting. I esp like the fact that
you can get a 'board' version for HALF the price?

The reason i was asking about the reso and how you can
apply it was because my first thought was to use the
device with the computer as a quick battery charger.
Having to measure the delta V of each cell, this sounds
like it could do it. Using a 1.2 volt reference diode
and setting to the 0.5 volt full scale i guess you
could get the resolution required to measure the
delta V accurately to determine the charge termination.

Im thinking of investing in some AA NiMH's but i can't see
spending $40 on a charger :-) and im not into monitoring
each cell with a meter while they charge on constant i.

Newbie:
Yes, i've looked into their chips, and even
designed a 100MSPS board to use with one of AD's chips.
The board stored something like 32k samples.
I guess now im looking for something complete and built
up already though.

Take care,
Al

 

[dat2zip]

Someone recommended the Labjack to me last year. I have one and it works for me. There are enough basic app interfaces to most simple setups.

I have used mine to do flashlight runtimes (light and temperature) together. Real nice for that. Output data isn't formatted the way I would like for direct import into a spreadsheet, but, a few global search and replaces will fix that.

I did another setup with a constant current load simulator. I measured the load current and tested some batteries not too long ago. That was a single channel and I forgot which tool interface I used but it had an equation section and I was able to multiply and offset the reading so the result was in the proper scale.

For the price, I would highly recommend it for a simple plug and play. Even though it's probably a single chip microprocesser with built in USB and A/D it's the software, drivers and interfaces that take a long time to develop. I can whip up the hardware in one or two days. But, doing the application, driver, firware and interface layers of software takes a long time.

Wayne

 

[Doug Owen]

[ QUOTE ]
dat2zip said:
Someone recommended the Labjack to me last year.

[/ QUOTE ]

Guilty, as charged. Didn't know you'd bought one, guess that's why you never took up the offer to borrow mine....

Glad it worked out fer ya.

Doug Owen