Driving 0.5km CAT5

What does "long" mean in this context? Milliseconds? Minutes?

What's the minimum such interval?

What delay can you tolerate between sending the pulse, and receiving it?

Sylvia.

You haven't told us anything about frequencies or time periods you are deailing with. That will pretty much dictate what's easily possible over

Standard differential RS485 for example is simple to implement and capable of more than 1km at low enough data rates.

Dave.

Sorry about omitting the info. I noticed it after posting my message. It will be of the order of minutes with a minimum of perhaps 1 minute.

Sorry about such a glaring omission. I'm talking intervals of minutes. Pulse duration is not yet decided but not critical in any case - may be of the order of a millisecond. Triggering is most likely to be with the rising edge. Exact triggering level is also not critical as long as rise time can be kept below 1 msec.

Pulse specifications (pulse width)? Rise time? Fall time? Pulse width? Minimum 'off' time? Maximum 'off' time? Signal level of pulse?

I provided most of the specs in reply to the others - in approximate terms since most of them are either not critical or not yet decided. Here's a summary:

Width - 1-10 msec Rise time - fraction of a msec at Tx, max 1 msec at Rx Fall time - not relevant as triggering will be by the rising edge, may be several msecs. Pulse width - milliseconds, variable. Single pulse. Minimum off time - 1 minute Max off time - indefinite Signal level - at least 4.5V. May also be from 12V logic circuit.

On a sunny day (Fri, 8 Jan 2010 15:57:40 +0530) it happened "pimpom" wrote in :

Risetime? Frequency? Amplitude? Pulse width? Electrical and magnetic noise in environment? Price. Delivery times? Availability?

As a cheap solution for low frequency and good noise immunity some pieces of rope will do.

All given before.

Car/bike ignition, cellphones and the like. No strong EMI expected. At least the cable will not pass through a microwave oven.

No stringent constraint.

No strict deadline. Preferably within a couple of weeks.

See my introductory lines. Preferably built out of common general-purpose components.

Try tugging on 500m of rope with less than 1msec delay at the other end.

On a sunny day (Fri, 8 Jan 2010 20:40:07 +0530) it happened "pimpom" wrote in :

There was no mention of 1ms in your post. For 1ms delay you need a risetime of 10x better. Your limit is about 10ns / meter for a signal to travel...

Ah, I have it, a big stroboscope flash, and a photo detector with lens pointed at the stroboscope :-) Or just use a coax. Or optical. Sound is too slow ... only 300m / second, so are pigeans. But pigeons can carry an SDcard or USB stick with a lot more data. LOL

How about small solid-state relays as the receivers? Simple, isolated, schmitt trigger action, slow enough to avoid reflection and crosstalk issues.

John

There was - in the follow-ups in reply to those who asked. (And I apologised for omitting important info). That's the point of a progressive discussion.

No problem with providing Your limit is about 10ns / meter for a signal to travel...

Ah, speed of light, with some degrading factors.

That may not be too far out an idea - for some future projects if not for this one.

Or just use a coax.

I considered coax before CAT5. But that would require either two cables or coding the two signals. That would mean a higher cost and/or complexity.

Semaphore? :-) See the third para of my initial post. I'd already checked out the feasibility of fiber. 2-core optical fiber costs less than coax or CAT5 and would be immune to interference. But the associated systems seem to obviate or reverse the cost advantage, and some elements may present problems of procurement.

With a magaphone?

:-)

We make these...

but they would be gross overkill at millisecond speeds. But multimode fiber-connectorized lasers and photodetectors are around $15 each and could be interfaced with simple circuits.

But I think CAT5 or equivalent and SSRs would be a safe way to go.

John

Standard RS-422/485 hardware should be capable of well over 200 kbit/s at 500 m. In addition, some manufacturers have transceivers that are slave rate limited to 250 kbit/s, thus reducing problems with reflections and sensitivity to external narrow EMI peaks. The rise and fall times are in the order of 1 us, so quite accurate synchronization could be achieved.

At such large distances, I very much doubt that the quite limited RS-422/485 common mode voltage range (-7/+12 V) would be sufficient in practice, especially considering lightning induced voltages, so the connection should use galvanic isolation at least at the other end. Typically the transceiver would be powered by a power supply and optoisolators would be used on the TTL side to connect to the rest of the station.

However, if isolation is required, why not use some DC-free coding such as Manchester coding and use small signal transformers (as in Ethernet) to isolate the reference levels from each other and from the line ?

I don't anticipate any problem at the receiving end as long as the pulse arrives in reasonably good condition (correct me if I've overlooked any potential pitfalls). What I need help with is in how the cable will affect the signal, estimate of immunity to outside interference, crosstalk between the two parallel twisted pairs, and any other factor I haven't thought of.

I've asked an associate to construct the circuit below, solely as a temporary test setup. Please comment.

I think I failed to consider the effects of the fairly high impedances with that simple test circuit.

.......................

Life would be much simpler if I could just order your products or other suitable ones. Unfortunately, the hassles involved with sending for something from another country make that option impracticable. I'm in a remote part of India and even ordering from another part of the country presents more hassles than it would for you to buy from China.

On a sunny day (Fri, 8 Jan 2010 23:33:14 +0530) it happened "pimpom" wrote in :

The 10k pull one way is weak, compared to the pull the otehr way by the transistors. For impulse it is much much simpler to wind a few turns on a suitable ferrite core to make a differential driver. And that then gives good DC isolation, as ormal with CAT5 interfaces.

On a sunny day (Fri, 8 Jan 2010 22:18:44 +0530) it happened "pimpom" wrote in :

Just beware of lightning flashes for false triggering :-)

Use a 600 ohm 1:1 audio transformer. Wire it like you'd drive a relay, e.g., open collector with reaction diode. Connect secondary to cable. Same thing on the other end. You will get a strong pulse at the recieving end, which can drive a B-E junction directly, or if you want to be fancy, use a hysteresis comparator. You might add some filtering to remove spikey trash.

In principle, driving DC through a transformer is stupid, and putting a diode across it is even stupider, but with minutes+ between pulses, it won't matter. Imagine it's an ideal transformer, ignore DC and be done. Just make sure it's got enough inductance, you don't want it saturating.

You get four twisted pairs in a CAT5, so you can send plenty of signals down the line. Or have signals going both ways.

Is CAT5 actually cheaper than two runs of ordinary twisted pair? Straight pair (zip cord, speaker cord, ...)?

Tim

-- Deep Friar: a very philosophical monk. Website: