EE's who do PCB Layout

I used to do that as well in the late 80's. But starting in the early

90's I found that the Internet and a good Gerber Viewer could replace such trips quite well. Never got any black eyes from that approach.

And no, Keith, I did not kill my layouter's cat ... But whenever I went over there for a layout check or a difficult initial placement it hopped into my lap. Which the layouter and I found astounding because I am a dog person and he said the cat would generally never do that with clients.

Fairchild NC7SV74. US8 package, single D-flop, 1 ns typical prop delay at 3.3 volts. 10K ECL wasn't that fast.

Works down to 0.9 volts, but a bit slower.

John

Hi Joerg,

I suspect you will be able to make this work, but unless it really is just a very occasional check, it sure seems as though hitting the road might be better overall; it's just so much faster to be able to position things yourself!

Yeah, understandable.

Hey, do you think you can design a fixed-frequency FSK receiver for ~900MHz (fixed frequency, no tunability required), some low kbps data date, and running off of ~3V (not too critical there) but 10-20mW (that is somewhat critical)? :-)

---Joel

Nice.

Something like 4-5 years ago when TinyLogic started to become popular, I lamented that -- at the time -- they were missing a single-bit 'flop and a single-bit tri-state buffer. I was happy to see when I looked some months ago now that both are available... and I've since used them as well!

True. But being there on a regular basis can be a bit tough if the client is on the other side of the Klondike river ;-)

If it doesn't have to be right now. Did you guys consider using the receive section in chips like this one?

20mW will be a challenge at 900MHz though, even with this chip. Would auto-polling be possible in your application or is it pretty much a continuous data stream? What can also help is to transmit the data in chunks at a much higher rate and send the receiver to sleep during breaks.

No, it's not right now... just starting to think about an upcoming project which appears to be about 90% of a sure-thing at this point.

I haven't looked at that one before; thanks for the link. (Some of the low-power parts that we've used in the past were aimed at, e.g., paging applications and unfortunately are now gone.)

It sleeps for awhile, wakes up to listen every now and again, and repeats. (It initially has to listen for awhile to "sync up" and then, knowing it's own ID, can figure out when to sleep and when to listen.)

---Joel

Good. Many of the calls I get are like "Well, the roof is already on fire...".

Yes, pagers are almost gone except for emergency crews. Even if the chips weren't gone I would not bank on them for anything that needs to remain in production for years. And from experience I'd try to stick to mainstream manufacturers and be careful with recently sold or acquired companies or product lines. Sometimes those parts can do a quick Houdini.

Take a look at chips for higher-end keyless entry systems. Meaning the ones that really work, not the super-cheap super-regen receivers.

If your 20mW spec is for the average consumption that should be no problem. If it is for the active phase it'll present a bit of a challenge at 900MHz. But I guess challenges are why guys like us chose engineering.

On another email list I'm on with UK members there's an expression:

"In the US they think a hundred years is a long time. In Europe they think a

100 miles is a long distance."

Robert

And EEyore thinks 100 is a high IQ. :(