A Jfet to use for switching?

The low pinch off suggests that you really would want one listed as a low noise one not as a switching one. Switching ones are usually optimized more for the RDS issue.

IF1331 from interfet

No chance of using an integrated CMOS analog switch?

John

I only need one switch, and as a hobbyist doing a prototype on breadboard I was hoping to use through-hole components. All the single switch CMOS ICs I've seen have been in surface-mount packages. I suppose if I have to I can overcome the aesthetic objections of using something like a 4066 for this application.

There are some cute, cheap single-gate CMOS switches, but they're SOT-23s or smaller.

John

You can also use a MOSFET output optocoupler in a 6-pin DIP. Here are some from NEC:

But you probably need a JFET for bidirectional analog switching. Try an H11F1M:

And for higher voltage (400V) try:

Paul

Try Vishay J201/SST201 or Fairchild J201/MMBF201

--
Thanks,
Fred.

I've used bidirectional mosfet SSRs as analog switches and they work fine. They are available in dip and surface mount.

NEC PS7141-1A is one we use like this. It will switch 400 volts! Offset voltages on these things is zero.

John

You really should try to overcome that limitation. If you shun SOT23 devices you are going to be missing out on a world of new and wonderful devices. BF862 and such.

[...]

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

Looking at the datasheet, I think this device will work nicely in my application.

Agreed. Small PCBs are inexpensive enough through services like BatchPCB that it seems silly to spend time constructing some circuits on protoboard. For more experimental designs I have been able to find some custom designed blob-boards that support both through-hole and surface mount components.

Thanks gentlemen, for all your helpful advice!

Until recently I used all THT components, even for production. But I found that, with a bit of practice and good magnifying goggles, good lighting, and a decent soldering iron, I was able to handle SMT components quite easily. SOT23 has 1.9 mm spacing, which is no sweat. SOIC components have

1.28 mm and are easy to work with. My greatest challenges have been PLCC devices with 0.95 mm and TSSOP with 0.65 mm pitch. That was pretty much the limit of what I could do with the tools I had. I had considered a 0.5 mm QFN-28 part but decided it was beyond what I could do by hand.

I find that I don't do a whole lot of breadboarding anymore. Most of my projects will eventually be used for at least small scale production, and I usually make the first prototype PCB with several alternative parts and room for extra components and thru-hole test points where I can hang other parts if required. I have some adapters for SMT that have pads brought out for breadboarding but I've seldom used them.

The technique I use for SMT soldering is to coat the pads with liquid flux. Then I set the SMT device in place, and the flux is usually thick enough to make it stick. Then I hold the part in place with a pointed probe or tip of sharp tweezers, and touch one lead with a slim tip soldering iron with a small amount of solder on it. When it melts, I remove heat, and then solder an opposite leg. With the right temperature iron, I find that you can solder the remaining legs easily, using minimal extra solder, as the board often has almost enough solder already plated on the pads.

Paul

With one clever cut you can place SOT23 parts onto almost any Vector board. Onto the ones with square pads even without cuts if you accept the odd looks of parts placed at 45 degree angle.

I use the thinnest Kester No-Clean there is, 15mils. Leaded, of course. Gets the job done and I don't have to clean up the flux.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

l

und

,

ve

C

the

m

nd I

her

out

lux.

h to

p of

a

lder

ard

Never use "no clean" flux. The stuff is just nasty because it only takes a little exposure to any hydrocarbon to make the ions get loose and start causing trouble.

Never use "no clean" flux. The stuff is just nasty because it only takes a little exposure to any hydrocarbon to make the ions get loose and start causing trouble.

----------------------------------------------------------------

I have a roll of no-clean flux but I have not used it. I don't like the way it works (or not), and it still leaves a visible residue. I have been using a spool of 60/40 0.025" wire solder that does not mention any flux, so I assume it is just solder. I use GC liquid solder flux and apply as needed. I clean the flux off the board with Isopropyl alcohol and a stiff horsehair acid brush, and then scrub it with spray detergent and a rinse with hot water. A final dry-off with a heat gun finishes the job.

I've had problems with ionic contaminants that cause leakage and noise in a humid environment, and I've seen a build-up of white crystalline deposits on the solder of boards that I have not cleaned properly, after sitting around for a few months. The ideal final touch is probably a rinse with deionized or distilled water and a thorough drying.

I've had mixed success with conformal coatings to reduce leakage in humid environments. Sometimes test equipment is stored in an outbuilding or transported in a truck and then brought inside where condensation sometimes occurs and there is almost no way to avoid some problems. But removing any trace of salts on the board is really important, and usually any leakage problems soon disappear. In some cases I have also added a power resistor as a heater to reduce and remove condensation .

Paul

On Feb 14, 7:58=A0pm, "Paul E. Schoen" wrote: [....]

es

y

The best solution is to put the board in an air tight metal housing. Plastic isn't good enough because the water sneaks through over time.

Conformal coatings are usually far more trouble than they are worth. They slow down the water getting into the PCB but they don't stop it so the PCB ends up seeing what the average of the environment is. If it is nearly always wet. That is what it will see.

To dry out boards, a few days in the hot box is needed. You need to allow enough time for the water that has made its way into the FR4 to get back out. Once the board is clean and dry, putting it directly into the housing means that it doesn't get time to absorb anything.