Question about RS232

I'm looking at a chart, right now, from Digital Equipment Corp, where they tested data rates vs distance with shielded and unshielded cables. Both cases used 22 AWG wire. In one case, it was 4-conductor quad (DEC part number 9105856-04, which I'm not going to try and look up as they are out of business and it's not worth trying.) In the other, it was two 22 AWG twisted pairs, each shielded (adds capacitance) in Belden 8777 (three pair.) The shields were tied to ground.

The table looks like this:

RATE Unshielded Shielded

---------------------------- 110 3000 ft. 5000 ft. 300 3000 ft. 5000 ft.

1200 3000 ft. 3000 ft. 2400 500 ft. 1000 ft. 4800 250 ft. 1000 ft. 9600 250 ft. 250 ft.

The cables were strung throughout office space and light industrial buildings and not kept on reel. Noise pickup from electrical machines and fluorescent lighting was small by comparison signal-to-signal within the same cable. The ever popular (here, anyway) 1488 drivers and 1489 receivers were used (not the 1489A, though), with a ground potential difference of less than 2V between receiver and transmitter stations.

The spec says that anything more than 2500pF, cable and receiver included, exceeds the spec. And the above table entries all exceeds the 50'/2500pF guidance.

The 1488 and 1489 were widely used, back then. I used them on my hobbyist stuff, despite the fact that the driver IC ran very hot to the touch. Wasn't a lot of choice for hobbyists at the time and they did work okay.

Jon

Out of curiosity, what bad experiences? One of our designs uses SP526 parts, and we've had some reliability problems in the past running in 485 mode, notably the inability to drive the termination resistors to proper voltage levels. Seems to have worked itself out now, though...

JT > I designed both the original flavor 1488's and 1489's (~mid '60's).

Man, Jim, the 1488's and 1489's were WIDELY used, for a LOT of years!

Has anybody ever told you what the production counts of those were??

You didn't get a royalty off those did you?

What was the toughest competitor for

1488/1489's for the first 40 years they were produced and used?

What other chips did you design that were produced in such numbers and for 50 years???

What was the problem? Had an (ST, I think) xx3232 fail for some reason (just one, out of one in use), which I've never seen with the MAX232/202 and similar parts.

What was the other one? One of their multi-digit LED drivers (for multimeters, etc.)? I seem to remember they had a ton of those things kicking around early on...

Problems when operating at elevated temperatures (180 out of 180 units failed after a few months), very susceptible to static discharge and just not working.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

ities.

That's true of VESA or DVD standards, but not EIA; my old copy of RS-232C (August 1969) bears a $5.10 price sticker.

That depends on the resistance of the lightning abatement system, and the current flow through it. Ohm's law sill applies.

--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

Well, I * R, obviously. But I don't think "lightning rods" are actually intended to divert a direct hit, although a substantial one might minimize the damage; their purpose is to drain off the space charge[1] that would lead a strike to hit your building.

[1] or whatever you call the static buildup that would make the air more prone to ionize.

Cheers! Rich

But 485 does not require those higher voltages. 485 is actually a low voltage scheme, down to 200 mv or something like that.

Jamie

... the RS-232C standard.

The most recent is TIA-232 F, October 2002, from

and the price has gone up to $142

kicking

Actually I have one of those meter boards laying around, but that wasn't the big seller. The 7660 charge pump was obscenely profitable. It may have been the case that Maxim had the body snatcher patented, though I see they now call it the ICL7660, so who knows.

Not silly. The 25V rating was to give protection to the equipmwnt, and not necessarily to the driver/receiver.

At last! A more complete dissertation of the spec without mandatory kilobuck fees! Also, actual voltage levels, *including* what a zero and one is, mark / space is!!! FUR out (sez the Baer).

It was the two 120 ohm terminators between the differential data lines at each end of the run. The driver was unable to maintain the 150mV differential voltage that the chip spec required.

res

owered

ecs.

t d

What you are proposing is a spec no decent semi would use. Think about it. How would you write up that spec. What is the test circuit for the DUT? It's OK to blow up the part, but only blah blah blah joules of energy is sent to the next chip? You don't even know the capabilities of the chips down the line.

Thus you spec the part to handle the voltage. Period end. When you sell something by the millions, you just can't screw around. A small failure rate could mean thousands of blow up parts coming back to the factory, with each customer bellyaching for FA.

I think that (the original( Fairchild was a rather decent semi company. And (like i said) they made triple line drivers and triple line receivers that (easily) passed all specs, INCLUDING the 25V spec. Required capabilities is IN the spec, so that "don't know" is hogwash. It takes no genius to make a circuit to pass that spec, what is required is the expertise to create the process for large volume at high reliability.

I designed in a 7660 last week, or suggested a client to do that. But I would never have done that sans 2nd source. We used one from National as

1st source. I think half a dozen companies make these.

--
Regards, Joerg

http://www.analogconsultants.com/

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

I don't know about the patent. I presume you mean by "body snatcher" that the pump pulls the NMOS-body/substrate down? I've seen similar stunts in Motorola/ON-Semi dual voltage chips ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

      Remember: Once you go over the hill, you pick up speed