[time-nuts] A few questions about Tboltmon
kb8tq at n1k.org
Fri Aug 14 08:13:53 EDT 2015
One might think that the dead antenna and blown GPSDO bias could be related….
One of the things that is probably worth repeating for the 100th time is the need
for a proper lightning arrestor on your antenna line. Both the antenna and the
GPSDO expect there to be a suppressor on the coax. It needs to properly
ground the shield *and* clamp the center conductor to shield voltage. Oddly
enough there are a lot of suitable arrestors on the auction sites. Even more
oddly, they often show up as part of a “cell site GPSDO installation kit”. The
arrestors can be had for < $20. Last time I saw the kits cheap they listed at
If you have ever been on a tower when a big cloud rolls over, you have heard
the thing “buzz”. That’s a strong suggestion that it’s time to get back to the earth in a
hurry. You don’t need to have a lightning strike to have significant DC sparking off
an elevated object. If you can hear the discharges they are *plenty* strong enough
to nuke a GPS antenna (if they discharge at the right point).
Many antennas are designed with an internal DC short. If you look at a common
low band VHF FM base antenna, it’s likely to be a Marconi self shorted design. Because
we feed DC to a GPS antenna, they don’t have this sort of self protection. Depending
on who designed the front end of your GPSDO, it may be pretty wide open as well.
The tendency is to think about this stuff in terms of “I never get hit by lightning”. That’s
generally true on any given day. I’ve had the house hit by lightning years ago. It’s
probably not a good guess over a lifetime. I have big clouds roll over the house
a few times a month in the summer. That’s a much more common thing. It does
not destroy lots of stuff. It can make odd things happen with elevated structures….
Buy the arrestor. Spend the extra $15 and get a good one from a brand name you have
heard of in relation to arrestors. Mount it as well as you can. Ground it as well as you can.
The electrical code in your area probably has some rules about this sort of thing. They
are well worth considering as well. Their objective is to keep your house from burning
down. That’s a worthy goal…having your house burn down *and* finding that your
insurance does not cover it could be a major bummer.
Buy the arrestor ….
> On Aug 13, 2015, at 9:57 PM, Chris Waldrup <kd4pbj at gmail.com> wrote:
> Thanks everyone. In addition to the blown transistor that feeds 5V down the coax, the one week out of box Motorola hockey puck antenna was blown too. I bought it in 2006 as a spare and I broke the shrink wrap two weeks ago.
> So I will order another spare but see if I can fix this spare.
> It's definitely worth it to have a few extras!!!
> Sent from Mailbox
> On Thu, Aug 13, 2015 at 8:01 PM, Magnus Danielson
> <magnus at rubidium.dyndns.org> wrote:
>> On 08/08/2015 07:05 PM, Bob Camp wrote:
>>> A factory reset will not brick the unit.
>>> 1) Your TBolt is blown
>>> 2) The cable has an issue
>>> 3) The antenna has an issue.
>> I've seen them all over the years, so neither is necessarily the most
>> likely. I'd also add:
>> 4) Power-issue
>> 5) Serial cable problem
>> For this case 1-3 should be your culprit.
>> Oh, do remember that engineers invent the most complex scenarios of what
>> the failure mode is, but fail to identify the simple ones such as power,
>> cables and connectors failing.
>>> For troubleshooting this sort of thing, multiples of each are a
>>> handy thing to have. Baring that:
>>> 0) Put a DVM on the coax and see if you have bias to the antenna
>>> 1) Hook up a TDR to the cable and ring it out both with a load and a short on the end.
>>> 2) Put the antenna on a *very good* spectrum analyzer and look at what is coming out.
>>> 3) Grab a signal generator that will simulate a GPS constellation and drive the TBolt with that.
>>> Since nobody (other than Magnus) ever has the sort of gear for 1-3, and it’s all pricey stuff
>> TDR is a nifty tool for this sort of thing so 1) is nice, but it won't
>> really help you and you will have to know what to expect from a
>> unpowered LNA. Spectrum analyzer will not directly help you since the
>> satellite signal spectrum is below the noise-floor, but you *might* see
>> the amplified noise as shaped by the LNA pre-filtering, which is
>> hopefully SAW filtered, so 2 is doable but tricky to interpret for the
>> novice. If you have a constellation simulator lying around, it will help
>> you to see if the receiver is working at all, but even I don't have that...
>> Having a VNA helps, and the nifty TinyVNA for instance will be quite
>> useful. Similar to the TDR, it sends a signal up the wire and analyze
>> the response, but in frequency plane rather than time-plane. Again, some
>> experience is required but this is a good time to learn.
>>> the simple answer is:
>>> 1) The antenna is probably the cheapest part of the setup. I’d swap it out first.
>>> 2) The cable is cheap but a pain to run, is it #2 or #3.
>>> 3) Hook up another timing receiver to the cable. There are lots of them out there in the $100 to $150 range.
>>> TBolts do die. My experience is that roughly 1 or 2 in 50 show up with a fatal issue. Another 1 or 2 show up
>>> with a (correctable) minor problem. I have had one drop dead after running for a while.
>> I would grab an antenna, toss it out a window and see if I get anything.
>> It is always handy to have additional antennas and cables around, and
>> for checking things to be operational, only modest requirements in type
>> and position is needed.
>> Similarly, having another GPS receiver to see if I get any form of
>> signal is a great tool.
>> Just taking the time to do quick and dirty tests helps. I've found that
>> I made stupid mistakes, so just doing a round of quick reality checks
>> have been important hints to find errors.
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
More information about the time-nuts