From: John Eric Nelson
jnelson@cmsa.gmr.com

Subject: SUMMARY: Portable data recorders

Thanks to all who responded to my post for infomation about portable data
recorders. The information I received was helpful and much more than I could
have found on my own.


__________________________________________________ ______________
My original post was as follows:

Subject: Portable data recorder?

Does anyone know of a portable/battery powered data recorder that could be worn
on a subject's belt. I need to be able to record multiple channels at 100 or
1000 hz and download the data at a later time. The device would be used in an
EMG related study. I have had no luck finding such a device so any help would
be appreciated.

Eric Nelson jnelson@cmsa.gmr.com
GM, NAO R&D Center, RTP, Bldg. 1-3
30500 Mound Road, Box 9055, Warren, MI 48090-9055
(810)986-1829 fax:(810)986-0294
__________________________________________________ ______________
__________________________________________________ ______________

The following are the replies that I received. I generally deleted the senders
personal information. If you would like to get in touch with the sender of one
of these relies please contact me and I will give you that info.


__________________________________________________ ______________

We have developped and we are using in our lab a static data recorder with
the following specifications:

8 or 16 analog channel inputs (selected by software)
0 to 5V or +/-5V (set with a switch)
12 bits +/-1 Analog to Digital Converter
25 Hz up to 1600 Hz (8 channels) or 800Hz (16 channels) sampling frequency
(selected by software).
The memory is size is 4 megabytes in our model by I know that it is possible
to get 16 megabyte model.
The maximal sampling time depends of number of channels, sampling frequency
an memory. For instance with 8 channels, 100 Hz, 4 MBytes, the maximal
sampling time is about 500 seconds.
The weight is 400 grams (including battery).

The data are transfered into a PC file throught the parallel interface
(Centronic type). The static recorder is also programmed before sampling
throught the parallel interface.

For more details on the methods (transducers, experiments ...) you can
contact me.

For more technical details on the static recorder you can fax to:
E.I.P.
Mr Regis BONNEFOY
Fax: (33) 77 39 91 07 (France)
__________________________________________________ ______________

This bit of info probably won't be very helpful to you, but here goes.
About one month ago, a woman named Cheryl Faye, from a company called
Digitrace, came to speak to us about a portable footswitch system.
Unfortunately, for both of us, their current system has only 2 channels.
Since no other information, besides footfalls, could be recorded, the
device did not seem to be very useful to us, especially since we have our
own footswitch system in house with kinematics, kinetics, and EMG
data collection available as well. Anyway, the company is located in
Massachusetts, but I do not recall exactly where or what their phone # is.
Good Luck!
__________________________________________________ ______________

In response to your request regarding a portable data recorder. I am
curious as to the application that you may be using this for. If it is
solely for EMG there are very good telemetry EMG systems that may suit your
need. That permits portability and does not require downloading later. We
have such a system. We purchased it from Konigsberg Instruments at 2000
Foothill Boulevard, Pasadena, CA 91107 Phone# (818) 449-0016. The contact
there is Bill Mills. He has supervised custom system development for
similary telemetry EMG systems for UCLA, Georgia Tech, Centinela Hospital in
LA, our company, and is currently developing a system for USC. I know
personnel at all of these facilities and all seem to be pleased with the
service, quality, and the price. Please contact Bill Mills for
specification information, etc. He is happy to work with you to develop a
system for your needs. They have also developed very custom instrumentation
for NASA.

If your system needs to be more flexible in terms of the type of sensors
used (strain gauges, transducers, accelerometers?) then you are in a similar
boat we are ... that is finding a portable system that can do these in the
budget of a small company. We have made some progress in finding market
available systems that do some of the things we need. One such system
(DataPak - I believe German built?) is similar to your description ... a
portable system that saves data on cards to be downloaded later. The
Biomechanics Lab at Penn State served as a Beta Test Site and some Ph.D.
students under Dick Nelson used the system on cross-country skiers and
golfers, and I think one or two channels may have been EMG compatible. I
believe that system is now being marketed through a company in the Colorado
Springs area called Peak Performance. You may be fimiliar with their name
as providers of high speed video motion analysis systems. With a call or
two I could find out the contact there for the DataPak system.

Sorry I couldn't provide many specifics off the top of my head, but I hope
this limited information helps some. If you have any more questions, please
contact me directly.
__________________________________________________ ______________

For a similar probem I am successfully using a
standard portable DAT-tape recorder by SONY. If you need more than two
channels, you can develop an analog or digital multiplexer, which cost you
a little bit of labour, but is very cost effective.
Proffessional portable data loggers are (at least for me) way out too
expensive.

>

I am using the TCD-D3 DAT recorder by Sony, but any other portable DAT
recorder should also do the job. This recorder is battery operated. A
fully charged battery results in 2 hrs recording time.
Nominal bandwidth 1hz to 20 khz. I modified the input stage, so that this
thing now handles frequencies down to 50 miliHz.
The Signal Nois ratio is about 90 dB, which frees me from the necessity of
frequent adjustment of the input level.
Nominal input Voltage is about 2 V rms maximum on the standard line input.
If you use the microphone input you can get as low as 1 mVrms. In this case
the SNR drops to about 60 dB.
I am generally using the line input and handle the analog pre-processing
externally with a couple of low-noise Op Amps.
You can purchase tapes between 45 mins and 120 mins recording time. The
TCD-D3 also offers a slow recording feature, which multiplies possible
recording time by a factor of 2. In this case, the bandwidth is limited to
15 khz and the digital encoding is nolinear.

Date is being downloaded to the PC via an A/D Converter Board along with
data aquisition software written by one of our students. Commercially
available data aquisition packages (such as View DAC) should also do this
job.
I am also feeding the recordet signals into a fourier analyzer for
additional data processing. This analyzer returns a plot in HPGL, which is
postprocessed with Corel Draw and then pasted into my documents.

A small warning: do not use a mini disk (MD) recorder or a digital audio tape
recorder (DCC). These recorders have a built in data reduction algorithm,
which is fine for music, but no good for stochastic bioelectrical signals.
I tried it once - the results were funny, but showed no association to the
original signal.
__________________________________________________ ______________

I do not know of any commercially available portable data acquisition
systems. However, several such systems have been developed by various
research institutions. At the UC Davis, Prof. Maury Hull and Jeff
Newmiller developed a system carried in a fanny pack and used for
skiing and cycling research. I also think the USOTC (Colorado Springs)
has some portably system, perhaps a telemetry system.
__________________________________________________ ______________

In response to your request for data logging hardware leads I have the
following:

Mega Electronics ME3000P - a four channel EMG data collection system with
1Meg on board ram which we have been using for 2-3 years with good
success in the automotive industry.

Premed as Physiometer-PHY400 - A similar system with position sensors,
made in Finland (fax +47 22637581). Untested by me so...

A Data logging system developped for a large scale research project in
Sweden (the stuff is not mass marketed but I hear it _is_ for sale in
some form). This system is geared to collect LOT's of data over long
periods of time. Again - I've only really seen the picture/demo so beware.
__________________________________________________ ______________

Hi, Try one of two sources:
1) the Medilog system by Ambulatory Monitoring. Unfortunately I don't have
their address handy. It was somewhere in New York state.
2) Ergonomic Concepts (Krishna Menon) 201-667-1052, has a portable system
that measures EMG and other measurements.
__________________________________________________ ______________

We use a portable TEAC 9-channel data recorder (cassette tapes downloaded
post data collection) to record free floating astronaut EMG during space
flight. It has the capability of sampling at the rate you need. If you'd
like more information, please respond or call me and we can talk more.
__________________________________________________ ______________

We found a similar problem and built one ourselves which we could
"clone" for youif it would work. The unit is based on a TMS 320
microprocessor, an 8 channel A/D front end and 512K of nonvolitile
memory. As configured, four channels are "tuned" for electro
myography and the other four are more general purpose. The
differences are quite minor gain and offset modifications. We
routinely can sample in the 100 to 1000 hz range, although the
]recording time is farily short if you want to save all the
data at 1000 hz. In our uses, we usually have the TMS 320
programmed to extract the "features" we want and just save those.

The logger communicates on a serial connection to a PC and we use
Microsoft Excel for simple data display.

Componet costs are about $1500/unit and we don't really have a
price yet.

If this looks to be of interest, let me know and I'll have one of
our more technical people get in touch with specs.
__________________________________________________ ______________

A company called TEAC manufacture a small portable data recorder
(2"x4"x8", or something like that), that saves data onto basically
ordinary cassette tapes. I have used it to record EMG data from as many
as nine channels. I think if I remember right that the frequency
range was up to 1300Hz, but this also was dependent on the amps used
inside the data recorder. But as always, this doesn't come cheap! The
data recorders themselves cost a bundle, and you also need a data
downloading unit, and of course your A-D board and software. However,
that was a couple of years back and the prices and/or technology may have
changed to you might want to look into it....
__________________________________________________ ______________

NASA uses such a device during a week long interview for potential
astronauts. To get more info you might want to call the NASA JSC ((Johnson
Space Center) Medical offices at (713) 483 7999. It is called a "holter
Monitor".
__________________________________________________ ______________

We have a device here at the UVA Auto Safety Lab that we use to
monitor 4 channels of data for temperature studies in
cadavers. The device is battery powered powered and can be
connected to your PC and the data periodically downloaded with
a very simple program. The device is the Pocket Logger made by
Pace Scientific. Joe Dobson is the sole owner/employee. His
telephone number in Charlotte N.C. is (704) 568-3691.
__________________________________________________ ______________

Last I checked TEAC made a 4 channel recorder that was compatible with
their bench model 7 channel instrumentation recorder. East coast
(or maybe new england) rep is Brian Kiley: 617-272-4550. I have
the 7 channel recorder and it has worked pretty well.
__________________________________________________ ______________

You may want to look up the TattleTale product line. It has
been a few years since I last looked up this information so I
do not have the address now. This company is sure to
have what you want, though. You can buy a variety of small
data recorders or just the circuit board.
__________________________________________________ ______________

A company by the name of BIOMATION markets a surface emg
system that has a portable unit that allows you to
collect sEMG for a short period of time for later
downloading. The person to contact is Dave Hanneson at
(613) 256-2821.
__________________________________________________ ______________

TEAC Corporation, Japan has a portable tape recorder (analogue data), maybe
8 ch. If I remember correct the type is MR-10. Its rather expensive and has no
final amplifiers.
TEAC Corp. has representatives in most countries.
__________________________________________________ ______________

there are several companies in Germany selling portable data recording
systems for EMG, HR, 1 V analog input etc. I don't know if this is of
any help for you, but maybe they have also addresses in the US. We use PARPORT
from PAR Electronic in Berlin, which is a 6 channel system (e.g. 4 EMG + 2 HR)
that records at max. 10 Hz (newer version works at max. 1kHz) and is restricted
as it already integrates the EMG (integration time for the EMG is 0,5 s). The
complete system costs about $20.000 and has DOS software.
medNATICs system offers multichannel telemetry, has a recording rate
of 1kHz, a 20 Mbyte flash memory card and Windows software. It costs
$17.000 + VAT for the complete system.

Addresses:

PAR Elektronik GmbH
Einemstr.9
D-10787 Berlin
Tel. +30 2139055
Fax +30 2138542

medNATIC GmbH
Harthauser Str. 21
D-81545 Muenchen
Tel. +89 6422958
Fax. +89 6422978

__________________________________________________ ______________

The LogIT data logging system is about the size you need. However, I
think its maximum sampling rate may be too low; my impression is that
100Hz would be as much as you would expect.

For information on the hardware, fax DCP Microdevelopments Ltd
(Cambridge, England) 0n +44 1480 830 534

For info.on hard programming of the device and downloading/analysis
software, fax SCC Research (Chelmsford, England) on +44 1245 250 934

In th UK, the device is also marketed by Fisons Scientific Equipment,
fax +44 1509 321 893.
__________________________________________________ ______________

We have done some good experience with Tattletales. Try the following
address:
ONSET Comp. Corp.
536 MacArthur Blvd.
P.O. Box 3450
Pocasset, MA 02559
FAX:508-563-9477
Tel:508-563-9000
__________________________________________________ ______________

I have had such a device made by Logger Techn. AB, Trelleborgveien 10, S-21435
Malmo, Sweden. The technology is based upon a "credit-card" PCMCIA memory card
which can be dumped into your PC at intervals. My cards are 20Mb each and can
hold 1 hour 20 min at a sampling frequency of 1500 Hz. When the card is full yo
u
just switch to another card. The only disadvantage is that the format is memory
mapped which is good for programming in C++ but needs to be converted in your
want your PC data in ASCII. The logger is so light you will not notice it's in
your belt. It can be used for EMG and othe data sources. I use is to store data
from a triaxial accellerometer.
__________________________________________________ ______________

Try Tattletale computers. They are very small and will do the rates you
need. I don't know about emg, though. You may have to build some
amplifiers for those.
__________________________________________________ ______________

How far from your site do you want to record this data? BTS carries their
TELEMG system,a telemetry system which sends the EMG data from a portable
patient unit to a receiver. I'm not really sure about the range that this has
,
but I can check on it if you are interested in this information.
__________________________________________________ ______________

I have been using a TEAC HR-30E Cassette FM data recorder for a project on
skiing biomechanics the past two years. It can record up to 7 channels of
data, although in fact one channel is used for noise reduction, and another
is a memo (voiced messages) channel.

It weighs about 400g, and is about 15 x 10 x5 cm in dimension. It uses a
TEAC cassette tape, standard size, to record an FM signal. It can use a
standard 9 volt battery, or can be powered with an external supply.

A larger TEAC "desktop" recorder is used to playback, and A-D the signal.

I use the recorder in a skier's "bum-bag" (hip bag), but I'm sure you could
secure it off a belt if the subject were not too vigorous. Cost a few years
ago was (I'm told) around AUS$8,000; I'm not sure if you can get them
anymore, or if so what their price would be.

The alternative that I'm looking at is a PDA (Hewlett Packard HP200LX) with
a National Instruments PCMCIA A-D card, which would be smaller, have less
noise, and save the A-D step. The PDA would be around $1,000, the card (I
believe) around $2,000. You may need to develop some software to interface
the card and PDA.
__________________________________________________ ______________
::::
Try Thought Technologies in West Chazy, NY.
__________________________________________________ ______________

Do you actually need to store the EMG? I just designed a nice little gadget
(an "EMG Horometer") that measures the time that the EMG is on and stores
it on a clock display. You simply read off the duration of EMG for a given
number of strides or a given test period (for upper limb use etc.)

I'd be willing to sell you one for $500.
__________________________________________________ ______________

For portable data recorders try:

Onset Computer Corp.
P.O. Box 3450
536 MacArthur Blvd.
Pocasset, MA, 02559
Tel: (508) 563-9000
Fax: (508) 563-9477.

I haven't dealt with them but understand they make a line of portable data
logging products. An article in the August 1995 issue of "Sensors" magazine
described the use of their loggers to measure the temperature in astronaut
gloves during extravehicular activities on a shuttle mission. They have a
web site which might include the article at http://www.sensorsmag.com.
__________________________________________________ ______________

We use the TEAC HR10/HR30 data recorder. The HR10 has 4 tracks (channels).
The HR30 which we use has 7 channels. Different types have different tape
speeds and hence different bandwidths for the data. Type E (dc to 1250 Hz,
45 mins) Type G (dc to 313, 3 hours), Type J (dc to 39 Hz, 24 Hrs).
Signal to noise is only 38 dB and drift +/- 1%. So you have to make sure you
use the full dynamic range of the record/playback system. By the way it is
only a recorder so you need their recorder/player type R71, R81 or MR 10/30.
__________________________________________________ ______________

We have a unit which will handle up to 1000 Hz/channel with upgradable storage
memory and from 2 to eight channels of EMG, It starts at $1995. It comes with
software for downloading to a PC at a later time. Or you can port it directly t
o
a serial printer. You can also create your own protocols on a PC and upload the
m
to the portable unit. It is amazingly small, only about 2.5 by 5 by 1 inch. You
can also use it for biofeedback, since it has a two column digital LCD display
for alpha-numeric or histogram displays,,, and audio. We'll soon have it
running on Signalysis, our multimedia windows data acquisition program.

. Rob Kall
FUTUREHEALTH, Inc.
3171 Rail Ave Trevose PA 19053, 215-364-4445, fax 215-364-4447
Smile@CIS.CompuServe.com
http://doc.web.compuserve.com:2000/homepages/smile (temporary beta site)
__________________________________________________ ______________

At the August ASB conference at Stanford University, I saw a Paromed
datalogger offered by Peak Performance Technologies. The unit has
eight EMG channels, apparently to 250 samples/sec. Storage is to a
PCMCIA card. Email address is peakinfo@peakperform.com. Phone
number is 303/799-8686.
__________________________________________________ ______________

Try talking to manufacturers of Holter monitors. These devices record
cardiac rhythms for periods of about 24 hours, on multiple channels.
__________________________________________________ ______________

The DATABEAR may meet your needs.
Here is the information on the manufacturer:

Langan Products
2660 California Street
San Francisco, CA 94115
(415) 398-7664
langan@eworld.com
__________________________________________________ ______________

In collaboration with Dr David Fyhrie at Henry Ford Hospital, Bone and Joint
Centre, we used a portable data recorder to record 3 channels of strain data
from in vivo gauges mounted on human tibiae. The recorder was small enough
to be carried while subjects undertook vigorous activities. I believe the
device came from one of the Unis close to Detroit. You can send Dave Fyhrie
a message at fyhrie@bjc.hfh.edu,to get details, or call him direct at Henry
Ford.
__________________________________________________ ______________

We are using a portable data recorder from Onset Computer
Corporation. The one we're using is the Tattletale model 5F. It
is programmable in BASIC (program in non-volatile RAM), so you
can program an entire protocol for data collection, including
downloading to PC. To get high sampling rates, part of the
program has to be in machine code, but it's not hard to do if
you're somewhat familiar with microprocessors. I can give you a
sample program if you want.

Their address is:
PO Box 1030, N. Falmouth, MA 02556, (508) 563-9000, fax (508)
563-9477.
__________________________________________________ ______________

Check the system at: http://www.oso.com/~ariel/
This system is used in NASA utilizing a Note-Book to collect up to 32
channels of EMG and Load Cells data through wireless transmitters.


Eric Nelson
GM, NAO R&D Center, RTP, Bldg. 1-3
30500 Mound Road, Box 9055, Warren, MI 48090-9055
(810)986-1829 fax:(810)986-0294