Solid State Relays - What spec's should I be looking at?

[Skimask]

Still working on my rudimentary curve tracer, which I guess should really be called an Analog Signature Analyzer...

Got the basic circuit up and running...A bunch of parts are gonna get swapped out for better parts once I get 'er all figured out.  A PIC with a DAC providing the waveform, TL084's amp'ing up the signal w/ variable gain using digital pots, going thru a series source resistance out to the D.U.T., signature signals coming back in, another couple of TL084's, another couple of PICs doing the ADC heavy lifting, feeding back to another PIC doing the control and display of everything.
Working good so far.  Still working on the programming for changing frequencies, voltages, etc.  Right now, it's a pile of push buttons, but that'll change since the LCD I'm using has a touch panel on it.

Sticking point is the series source resistance between the opamp providing the wave and the D.U.T.  Right now, I'm manually switching the source resistor out by hand.  Tried finding a digital pot capable of running + and - at both end of the wipers/resistors and be able to carry 500mA at most.  No dice.  So, it looks like I'm going to be stuck with relays for switching out a variety of source current limiting resistors for the D.U.T.

Solid state relays - Can I assume they don't provide a "zero" resistance across the terminals when activated?  Do they pass AC waves down in the millivolt range?
As the title suggests, what specifications in the datasheets should I be looking at when I get to selecting an SSR for this?  Or is an SSR just not going to work in this application?

[Kremmen]

SSRs are designed to switch power, not signals. Their datasheets are not going to have too many useful details re this kind of application. Usually there is a triac inside and that will turn off at some low voltage across the device. Definitely not designed for millivolts. I would not consider one of those.
A fet is effectively a voltage controlled resistor, although achieving constant resistance over a range of drain-source voltages might be nontrivial. Operating in the ohmic region will be resistor-like however, so maybe that is something you could check. A link in point: http://freespace.virgin.net/ljmayes.mal/comp/vcr.htm

[ptricks]

This is one of those cases where a transistor would be better than a fet. Fets are great for lots of things but when it comes to performing current limiting a BJT is much easier to implement.

[ptricks]

Agreed it is typically simplier. I dont know exactly what it is he is trying to do. If he wants to just switch out resistors or what? What voltages are available etc...?

He is building a curve tracer where you need to use different currents depending on what you are testing, like in this circuit:
http://www.techlib.com/electronics/curvetrace.html

The main difference is he is not connecting to a scope but to an ADC to process the information.

Another option is a op amp that can do high current outputs.

[Skimask]

Aye...  A load of information, left out, I did...

SSR's - power vs signal - That's kinda what I figured.  Get down below a few hundred mV and the thing won't trigger.

Operating in the ohmic region would likely be close enough as long as said region doesn't change much, e.g. less than an order of magnitude at most.

Don't want to use mechanical relays for that reason there...mechanical.  Moving parts fail, and they're noisy.  If you've ever used a Huntron Tracker 2000, that's the noise I'm talking about.  Relays switching constantly as you're going thru the ranges and watching the signature changes.  (Re: resistance in a relay...Yes, true, but that resistance is practically constant and also low compared to something like a 4066)

FET switch - the page says the signal being switch on/off has to be greater than the Vgs.  I'll be pushing a source signal that can run down to ~200mV, maybe less if I can get away with it.  And using a transistor...wouldn't I have to contend with a diode drop?

OpAmp solution - hmmmm...are you saying use an opamp for each resistor "channel"?  Using it as a sort of switch?

Hmmm....thinking...thinking...thinking....
I might be better off going with a bank of relays in the early versions and work my way up to a fully non-mechanical method...
I ain't that smart yet

[Skimask]

If you really don’t want to use electromechanical relays for whatever reason you can make a bidirectional switch out of 2 FETS (N or P channel).
for a P-channel example SEE .

http://www.electronic-products-design.com/geek-area/electronics/mosfets/using-mosfets-as-general-switches

You can get 2 mosfets in a single package to keep the parts count down.

They will have some leakage but seeing as to how you are talking mV signal levels 20-60VDS fet leakage at those levels should be negligible particularly if they aren’t dissipating significant power to heat up. I'm not 100% sure what it is you are switching but thats how you make a bidirectional switch anyway.

Every electro mechanical or solid state switch has resistance no one has developed a superconductor. They are close at extremely low temperatures but not yet.

I'm back on the subject of trying to make an SSR that'll handle AC signals down into the mV range at mA's.
Regarding the circuit described in the link above, wouldn't the intrinsic diode in each mosfet add a diode drop to the signal itself?  Thereby rendering the signal useless at voltages below the Vf of that diode?

[houdini]

Every electro mechanical or solid state switch has resistance no one has developed a superconductor. They are close at extremely low temperatures but not yet.

Yes the have.  You cant buy a super conducting relay but you can buy a super conducting electromagnet.

[Skimask]

No you have two mosfets so they provide bidirectional blocking switch OFF. When you want to pass the signal they are both “ON” so the body diode is shorted leaving just IRdson x2 drop.

Ok, so I take a couple of 2N7000's (only because I've got a pile on hand), back-to-back as described in the link (although the link shows P-channel, shouldn't make much difference other than n-channel's are a bit better overall as far as threshold goes), give them a V_gs close to or slightly higher than the highest V_ds I'll use and should get 2x R_ds-on across them.
I'll give it a shot and see what smoke I can let out! 

[Skimask]

If you use NFET’s to turn them on the gate has to be pulled above the source by at least 5V to fully enhance it. Then to ensure it stays off you can’t let your AC signal pull the source below the gate. Then its the reverse case for PFET's. This may or may not be a problem depending on the magnitude of the signal and what voltage sources you have available to drive the FET’S.

which is the issue because I'm actually trying for an AC signal (positive and negative swing), not just a varying DC signal (always stays above or below ground), and furthermore, that signal will range from +/-15pk-to-pk down to the mV range.

Relays - That's what I'm trying to stay away from...the mechanical switch.  But if I gotta, I gotta...

[brainwash]

There's a nice presentation on youtube from w2aew showing how to use diodes as a switch. Granted, it is only good for high-frequency AC signals but it might help you look from a different perspective.

[Skimask]

Saw that video ~few days ago.  Good info...but can't apply it here.  Well, I could, but it would mean reworking the signal path to better handle low amplitudes...e.g. better components for less noise, etc.  I'm not up to that point yet.

[Skimask]

Have settled on an ADG1406 16 channel analog MUX....I think.  Want to run it by anybody that is probably more in the know than me...or cares for that matter
This is for the curve tracer project I'm playing around with.  Need to switch a select resistor into the D.U.T. part of the circuit.
The signal going into any of the inputs will run anywhere from 100mV to ~12v pk-pk, AC sine wave, roughly 20hz up to 100khz at most.  Resistors to be switched will run anywhere from 10ohms up to ~100k.  At the higher pk-pk input voltage settings, the lower value resistors will be 'locked out' so the chip doesn't get overloaded with too much current.

28 pin TSSOP, dual supply @ +15v/-15v (I'll be running it at +12/-12 at most), fairly low R_on at the switches, low enough for my app.
Page 7 of the datasheet (http://www.analog.com/static/imported-files/data_sheets/ADG1406_1407.pdf) shows it'll do 180mA per channel continuous, which is about double what I'll be feeding it, and even then, I'll be feeding a sine wave thru it, so shouldn't be hitting anywhere near the max current limit for any length of time.