Analog To USB

I'm a fan of LabJacks for that.

Cheers

Phil Hobbs

I have used those when my company paid but I asked for inexpensive since I am paying.

--- news://freenews.netfront.net/ - complaints: snipped-for-privacy@netfront.net ---

Ahh yes. I spend other peole's money alot faster as well...

How much bandwidth you need ? Maybe one of the cheap USB scopes would work. I think I saw some that were two channels.

Looks like those Labjacks are like eight channels and more. Looked like the one was around a hundred bucks. For eight channels that doens't seem so bad but I see that if you do not need it, why pay for it ?

Or, maybe there are used ones on eBay ?

They come down to about $75. What do you mean by "inexpensive"?

Cheers

Phil Hobbs

Get a USB sound dongle and remove the DC blocking caps? With that, you can use the windows standard media API's or that for 42 bucks.

Jamie

I have had good results with the HDE 5.1 USB device (about $15 on Amazon), though I used an additional diff-amp circuit on the input: . The units use a CM6206 chip and have a nice metal case with room inside for a tiny PC board to hold the diff-amp. (There are other identical-looking devices that apparently don't have the same internal layout.)

If you don't use a diff-amp, just bypass the caps, the digital output would be 'unipolar' with bipolar encoding: zero input would be reported as negative full-scale, more-positive inputs would be reported as less-negative, up through zero to positive full-scale.

The link includes schematics and photos for this device, but you can probably do the same basic diff-amp mod to most USB sound devices. You just have to figure out what ADC chip they use and find its Vref pin to use as the diff-amp ref. Then you just trace the input channels to find the input caps and remove them, and connect your diff-amp between the former cap holes.

You can use the same diff-amp circuit, which includes a link to a PCB layout you can print off, and one you can send out to have made by PCB Express.

As for the underlying pre-mod performance of this device, full measurements of this and several other USB devices are here:

Best regards,

Bob Masta DAQARTA v7.50 Data AcQuisition And Real-Time Analysis

Why not just use a common drain JFET, or, if you have enough headroom in the supply, a simple Zener with a bypass and a pullup ?

If you need alot of accuracy, something off the shelf would be better, even though it is more money.

Sorry I didn't specify - to fix the DC offset problem.

Not sure what you have in mind. The need is to take (typically) +/-2.5V signals from the external world and convert to 0-5V DC for the ADC, with the external ground referenced to Vref of the ADC. The diff-amp circuit uses a

(Actually, the ADC chips I've seen use 3V or 3.3V supplies, cut down from the 5V USB, with Vref in the 1.2-1.5V range.)

Best regards,

Bob Masta DAQARTA v7.50 Data AcQuisition And Real-Time Analysis

If you take an N channel JFET and put positive DC on the drain and zero vol ts (quiescent) on the gate, there will be a positive voltage on the source. Not like a MOSFET, it can shift the DC lavel this way in the "source follw er" configuration, a term I do not like to use because it is not right. Not hing follows the source, the source follows the gate.

Actually JFETs are a bit hard to get compared to other devices. they are av ailable but if you go to Digikey and compare how many types they have versu s other kinds of transsistors, there aren't that many. Many of them are use d for switching because of some other charsacteristics, but not all.

you just put the gate at zero volts with a resistor, put the drain to +5, a nd pick a resistor that gives you the right voltage at the source. Of cours e you have to keep the current and therefore the dissipation within limits but I see no problem with that. If the output impedance is too high you can use an "emitter follower" after it, but I really doubt that'll be necessar y.

I don't claim this will be totally accurate. The deal with the OP AMPs is p robably more suited for instrumentation purposes, but the OP wants cheap an d we don't really know the app at the moment. Actually, with a couple of mo re components ant drift can be compensated. A simple current csource from s ource to ground could be regulated to keep the voltage shift constant with temperature. I oculd probavbly draw someting up, but later because I am out ta here in a few, got some things to do. It really is simple, what I have i n mind anyway. The worst it would be is a JFET, a bipolar and a few resisto rs.

Oops, couple of diodes maybe.