[time-nuts] Can I get 1 millisecond accuracy with a USB GPS-18

[lists at lazygranch.com]

I think the jitter was around a microsecond. Hardly significant for the intended comm application. I designed 3 of those chips, but not the original MAX232. And this is over a decade ago, so I can't really be sure of the limit. However, I am positive the problem exists.

The interface chips had no specification on jitter. But when you test a driver or receiver on ATE, you need to set a limit on the maximum prop delay. That is, if the chip has a defect, the pin will never propagate the signal! You have to stop looking for a transition at some point and reject the chip or die. Test engineering needs a limit, even if the spec is internal. And if you set a limit, it has to be justified. Dry labbing is frowned upon. ;-)

So the parts would be bench tested over temp to brick wall this limit. A storage scope is set up to get the envelope. You'd run the storage scope for minute, so probably a million pulses were used to produce the limit. 

For short distances, you don't need the full 232 swing. You can roll your own transmitters and receivers with inverters, resistors and diodes.  I can't say I ever bothered to roll one out of discrete transistors. 



------Original Message------
From: Chris Albertson
To: lists at lazygranch.com
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Can I get 1 millisecond accuracy with a USB GPS-18
Sent: May 13, 2013 11:13 PM

On Mon, May 13, 2013 at 7:52 PM,  <lists at lazygranch.com> wrote:
> Note that MAX232 type chips add jitter. The charge pumps make substrate noise and that leaks into the receivers/transmitters.

How much noise or jitter?  Has anyone measured the effect?  Is it at
the nano or micro second level?   It would be easy enough to use a
transistor if it improved the timing enough to matter.

--

Chris Albertson
Redondo Beach, California