Out of the several hobbyist-affordable time sources to obtain (Garmins or any NMEA-output), I do think the Motorola Oncore UT modules are pretty top-notch in terms of reliability, stability and performance. I think it provides just the right amount of challenge and barrier-to-entry for anyone who wants to dabble in EE and do some TTL-to-RS232 logic shifting to get one of these to work.
Reasoning Behind This
1) Arrangement and un-honkify my setup: To nicely arrange the Motorola Oncore module with a MAX232 and a battery for backup onto some sort of prototype perf board. For the LONGEST time (3+ years), I had an individual TTL-to-RS232 DB9 breakout screwed to some wood, and used a DB9-serial ribbon cable (since the Motorola Oncoreh has a 10-pin male header and the DB9-serial conveniently has a 10-pin female socket + ribbon). I'm no engineer, but I like to pretend to be one on my spare time (and at work when I'm allowed to) but I thought it was about time to make my stuff look a little bit more professional and less honky.2) No onboard battery for almanac storing with power-loss: My UT+ modules do NOT come with an onboard battery, thus every time I'd lose power, the cold-start time on the GPS was enough to piss me off from time to time. I used to have a UPS but kW usage over time really added unnecessary 'cost' to my electric bill to the point where I could really live without forking over several hundreds of dollars to 'the man' a year for the few times I lose power intermittently.
3) Constant falseticker issues with current hardware setup and PPS from Motorola: For whatever reason (and I think partly due to newer ntpd versions, see this post) ntpd would label my PPS timesource as a falseticker all the time and it was frustrating. I didn't want to do some configuration hack and force truechimer status with 'true' in the ntp.conf config. What I was currently doing is sending the 1PPS TTL signal straight to DB-9 Pin 1 (carrier Detect) on the assumption that RS-232 logic is 3-24v, so clearly +5v TTL would 'barely' fall into it that logic path, but would. After some forum posting, I decided to move the 1PPS signal through the MAX232 TTL-to-RS232 shifter as well and my current circuit only took two inputs; one for Rx and Tx. So a redesign was needed.
4) Get the RS232 DB-9 serial output and antenna to be in the same location: Right now I plug my DB-9 serial cable in on the right side then the antenna is on the opposite side so it makes positioning this thing in my closet hard with my crappy setup.
The 'Solution' (so far)
1) Soldering time --- take a while to solder all that together when it's thru-hole.
2) DESIGN!!!: Because I wanted this to 'snap' together and sit on the underside of the UT module, I had to really think about this. I certainly didn't plan it out as nice as I turned out (I think). Here's another few pictures of me putting it together, along with the final product, and how it fits together underneath the Motorola Oncore module:
Parts List
Since I did my schematic with Digikey's scheme-it, I was able to make a nice bill of materials in case someone else wants to dupe this. I'm not endorsing Digikey, but you'll be able to see what I used; I had all these parts on hand from years of accumulation.Object Name Value Digikey Part #
R1 RESISTOR 1k8 CF14JT1K80CT-ND
R3 RESISTOR 470k CF12JT470KCT-ND
R4 RESISTOR 330k CF18JT330KCT-ND
J2 2X5 952-1924-ND
J3 9-PIN, FEMALE A32510-ND
IC1 IC MAX232CPE MAX232CPE-ND
C1 NON 1uf P4537TB-ND
C2 NON 1uf P4537TB-ND
BT1 BATTERY HOLDER CR2032 BU2032SM-HD-GCT-ND
J4 1X2 TMS-104-02-S-S-RA-ND
C3 NON 1uf P4537TB-ND
C4 NON 1uf P4537TB-ND
R2 RESISTOR 1k8 CF14JT1K80CT-ND
L1 LED Red 751-1089-ND
L2 LED Green 51-1105-ND
Get whatever type of thru-hole prototype perf board you want; I had some leftover stock I picked up from Radioshack a long time ago. Improvements To Add
1) Put some inline diodes for some circuit protection along with a 5v voltage regulator --- I'm under some wild assumptions that I'm the only one who will ever use this and always apply 5v DC supply to it.2) Still work on that battery indicator just because after I stuff this up in my closet, it gets easily forgotten. I stumbled across this for a 3v low battery flasher using a Panasonic MN13811-G, but that MN13811 was pretty hard to find anywhere (digikey, mouser, etc.)
3) Get it off the wood 1x4 mount and get a proper project box to enclose this.
NTPd info
Here's a snapshot of my tuned ntp.conf for the Motorla Oncore UT's (in case you need a jumpstart on it):enable pps pps /dev/oncore.pps.0 hardpps # Oncore Motorola UT+ server 127.127.30.0 minpoll 6 prefer fudge 127.127.30.0 time1 0.1988 # Oncore Motorola UT+ 1PPS on DCD server 127.127.22.0 minpoll 6 prefer fudge 127.127.22.0 time1 0.39726 tos mindist 0.010Here's a look at my peers listing:
$ ntpq -p remote refid st t when poll reach delay offset jitter ====================================================================== LOCAL(0) .LOCL. 8 l 486 64 200 0.000 0.000 0.000 *GPS_ONCORE(0) .GPS. 0 l 3 16 377 0.000 0.010 0.003 oPPS(0) .PPS. 0 l 34 64 377 0.000 0.099 0.133 $
Resources
A few sites of interest I referenced and used for the idea:1) TAPR Motorola Oncore Interface SemiKit (http://www.tapr.org/gps_vpib.html): I'm sure I've looked at this while re-referencing the Motorola Oncore manuals from here, but this is EXACTLY what I wanted to do. However, I'm not going to pay to get boards made, nor do I want to have an all-fitting Motorola Oncore interface board; I just have UT's.
2) Simple low battery indicators (http://kyorune.com/modding/article.php?id=55): This is very, very barebones. However, it gave me a good opportunity to do some testing, even though I kind of knew a head of time this was going to be too expensive milli-amp wise on the battery to even make it useful. The Motorola draws in micro-amps, so the the added value of an LED drawing at milli-amps just makes no sense.
No comments:
Post a Comment