GC-1000 and Arduino
92 views
Skip to first unread message
Mitch
Jan 31, 2012, 7:07:18 PM1/31/12
to neonixie-l
Hi All,
I posted this in the Heathkit clocks forum too.
I have three GC-1000s and living in South Florida, I've never had one
work well.
I've tried all kinds of antennas over the last few years with my
original,
without much luck. I have not tried an outside antenna, it's not
realistic
around here.
I decided to attempt to interface an Arduino so that I can connect a
GPS
receiver and at least have one of them working reliably. I realize
that this
changes the character of the GC-1000 to not much more than a display,
but that's
fine with me if I can have one work reliably. No permanent mods will
be made to
the GC-1000.
The hardware connection was made by lifting pin 8 from each of the
567s on the
tone board, connecting the traces to two digital output pins on the
Arduino.
I can send IRIG H001 frames to the pin that was connected to U402 from
the 100Hz
source, and the data light blinks appropriately. What goes to U403
from the
1000Hz source is unclear in the GC-1000 manual. I've tried sending P0
and also
the reference marker separately, and together. Sending the reference
marker during the 1 second quiet period at the beginning of the frame
seems to be correct. The capture light does turn on,
but time is never displayed. I don't know how long it should take, but
it's been
left on for a couple hours.
From the schematic it appears that the data coming from the 567s is
inverted, so
I've allowed for that in the sketch. I was sending the same frame
without
changing the time for testing purposes. Possibly that is a problem,
I'm not
sure, so I now have it send four frames each incremented by one
minute, but no change.
At this point I think I'm stuck. Has anyone done this, and can anyone
help?
Thanks in advance. Mitch
I posted this in the Heathkit clocks forum too.
I have three GC-1000s and living in South Florida, I've never had one
work well.
I've tried all kinds of antennas over the last few years with my
original,
without much luck. I have not tried an outside antenna, it's not
realistic
around here.
I decided to attempt to interface an Arduino so that I can connect a
GPS
receiver and at least have one of them working reliably. I realize
that this
changes the character of the GC-1000 to not much more than a display,
but that's
fine with me if I can have one work reliably. No permanent mods will
be made to
the GC-1000.
The hardware connection was made by lifting pin 8 from each of the
567s on the
tone board, connecting the traces to two digital output pins on the
Arduino.
I can send IRIG H001 frames to the pin that was connected to U402 from
the 100Hz
source, and the data light blinks appropriately. What goes to U403
from the
1000Hz source is unclear in the GC-1000 manual. I've tried sending P0
and also
the reference marker separately, and together. Sending the reference
marker during the 1 second quiet period at the beginning of the frame
seems to be correct. The capture light does turn on,
but time is never displayed. I don't know how long it should take, but
it's been
left on for a couple hours.
From the schematic it appears that the data coming from the 567s is
inverted, so
I've allowed for that in the sketch. I was sending the same frame
without
changing the time for testing purposes. Possibly that is a problem,
I'm not
sure, so I now have it send four frames each incremented by one
minute, but no change.
At this point I think I'm stuck. Has anyone done this, and can anyone
help?
Thanks in advance. Mitch
Cobra007
Jan 31, 2012, 8:18:20 PM1/31/12
to neonixie-l
I think this will be of some help:
http://www.nist.gov/pml/div688/grp40/wwv_format.cfm
You do need both 100Hz and 1000Hz signals and they should also come at
synchronized moments seems like.
Michel
http://www.nist.gov/pml/div688/grp40/wwv_format.cfm
You do need both 100Hz and 1000Hz signals and they should also come at
synchronized moments seems like.
Michel
John Rehwinkel
Jan 31, 2012, 10:26:57 PM1/31/12
to neoni...@googlegroups.com
> I can send IRIG H001 frames to the pin that was connected to U402 from
> the 100Hz
> source, and the data light blinks appropriately. What goes to U403
> from the
> 1000Hz source is unclear in the GC-1000 manual. I've tried sending P0
> and also
> the reference marker separately, and together. Sending the reference
> marker during the 1 second quiet period at the beginning of the frame
> seems to be correct. The capture light does turn on,
> but time is never displayed. I don't know how long it should take, but
> it's been
> left on for a couple hours.
> the 100Hz
> source, and the data light blinks appropriately. What goes to U403
> from the
> 1000Hz source is unclear in the GC-1000 manual. I've tried sending P0
> and also
> the reference marker separately, and together. Sending the reference
> marker during the 1 second quiet period at the beginning of the frame
> seems to be correct. The capture light does turn on,
> but time is never displayed. I don't know how long it should take, but
> it's been
> left on for a couple hours.
You'll need to send the modified IRIG used by WWV (17 cycles/170ms for a binary 0, 47 cycles/470ms for a 1). You'll also need to include the position identifiers every ten minutes (77 cycles/770 ms). The GC-1000 firmware looks for two consecutive frames that differ by exactly one minute before accepting the data as valid. It requires three of these pairs in a row before it will update its time.
> From the schematic it appears that the data coming from the 567s is
> inverted, so
> I've allowed for that in the sketch. I was sending the same frame
> without
> changing the time for testing purposes. Possibly that is a problem,
> I'm not
> sure, so I now have it send four frames each incremented by one
> minute, but no change.
Close, but I think you'll need six frames, each incremented by one minute.
> At this point I think I'm stuck. Has anyone done this, and can anyone
> help?
I haven't done this, but reading the GC-1000 manual (Theory of Operation section) yielded the information I mentioned above. I downloaded a copy in order to do so.
- John
Cobra007
Jan 31, 2012, 10:30:11 PM1/31/12
to neonixie-l
From wiki:
http://en.wikipedia.org/wiki/WWV_(radio_station)
I was thinking, once you got it running, you could actually make a low
power RF transmitter, modulate it with your signal, and use it for all
3 of your clocks at the same time without any modification to the
clock.
Just an idea
Michel
http://en.wikipedia.org/wiki/WWV_(radio_station)
I was thinking, once you got it running, you could actually make a low
power RF transmitter, modulate it with your signal, and use it for all
3 of your clocks at the same time without any modification to the
clock.
Just an idea
Michel
Mitch
Feb 1, 2012, 12:24:23 PM2/1/12
to neoni...@googlegroups.com
Works! Thanks for the help and comments.
It didn't work yesterday because I used a data array from a simple Arduino sketch that I found on a forum. Most of the elements were in the wrong positions, maybe it was for a different IRIG spec. It doesn't seem to matter if the first frame at 0 seconds (the hole) is 1.03s, not sure why they did that because it shifts everything else forward by 30 milliseconds. The GC-1000 syncs equally well with and without that delay.
It takes four data frames to set the time. Updating the time took five frames the one time I checked, not sure why. Maybe it was because the same frames were repeated and the time difference was so great.
Next I'll work with a GPS receiver and the Arduino's C++ Time.h library to get everything working. I will use an old RS-232 connector to hold both jacks for the GPS, and also for a mini-usb connector so that the Arduino sketch can be updated without opening the case. Only four wires will be connected to the clock, and the modification is easily reversible.
A low power WWV transmitter would be great. That's beyond my skill level, so this will suffice unless someone else designs that part.
Mitch
It didn't work yesterday because I used a data array from a simple Arduino sketch that I found on a forum. Most of the elements were in the wrong positions, maybe it was for a different IRIG spec. It doesn't seem to matter if the first frame at 0 seconds (the hole) is 1.03s, not sure why they did that because it shifts everything else forward by 30 milliseconds. The GC-1000 syncs equally well with and without that delay.
It takes four data frames to set the time. Updating the time took five frames the one time I checked, not sure why. Maybe it was because the same frames were repeated and the time difference was so great.
Next I'll work with a GPS receiver and the Arduino's C++ Time.h library to get everything working. I will use an old RS-232 connector to hold both jacks for the GPS, and also for a mini-usb connector so that the Arduino sketch can be updated without opening the case. Only four wires will be connected to the clock, and the modification is easily reversible.
A low power WWV transmitter would be great. That's beyond my skill level, so this will suffice unless someone else designs that part.
Mitch
Cobra007
Feb 1, 2012, 4:41:13 PM2/1/12
to neonixie-l
If I were you I would spend a little bit of time to work out how a
simple WWV transmitter can be made. It really is only a small step to
go from the point where you are now. It's a beautiful clock and if you
have a GPS to WWV converter (RF), the clock can be used globally and
is still doing the thing it was designed for, it's just that it is a
local WWV station :-).
They probably use AM or SSB, I drafted a (very) simple schematic just
to show you how simple this can be done. Sure you will have to work
out the rest of it, but this can be used as a very basic AM WWV
transmitter. SSB would be quite simple as well by multiplexing the
100Hz and 1000Hz tones at the carrier frequency, effectively omitting
the carrier itself this way.
http://xiac.com/Images/WWV_AM.jpg
Michel
simple WWV transmitter can be made. It really is only a small step to
go from the point where you are now. It's a beautiful clock and if you
have a GPS to WWV converter (RF), the clock can be used globally and
is still doing the thing it was designed for, it's just that it is a
local WWV station :-).
They probably use AM or SSB, I drafted a (very) simple schematic just
to show you how simple this can be done. Sure you will have to work
out the rest of it, but this can be used as a very basic AM WWV
transmitter. SSB would be quite simple as well by multiplexing the
100Hz and 1000Hz tones at the carrier frequency, effectively omitting
the carrier itself this way.
http://xiac.com/Images/WWV_AM.jpg
Michel
Mitch
Feb 1, 2012, 10:23:21 PM2/1/12
to neonixie-l
A 5 Mhz AM transmitter that can modulate 100Hz and 1000Hz, putting out
a few milliwatts, would be perfect. I know little about analog
electronics. Someone else will have to do that part.
a few milliwatts, would be perfect. I know little about analog
electronics. Someone else will have to do that part.
GastonP
Feb 2, 2012, 8:56:45 AM2/2/12
to neonixie-l
I would advise to be *very* careful on this specific topic of mini-
transmitters.
If that frequency is not a radio amateur one which does not need a
license to operate, it could be illegal.
I doubt much that a frequency used by WWV is open for general public
to stomp on Government issued transmissions so... check your local
regulations to avoid nasty surprises.
Gaston
On Feb 2, 12:23 am, Mitch <mitc...@gmail.com> wrote:
> A 5 Mhz AM transmitter that can modulate 100Hz and 1000Hz, putting out
> a few milliwatts, would be perfect. I know little about analog
> electronics. Someone else will have to do that part.
>
>
>
transmitters.
If that frequency is not a radio amateur one which does not need a
license to operate, it could be illegal.
I doubt much that a frequency used by WWV is open for general public
to stomp on Government issued transmissions so... check your local
regulations to avoid nasty surprises.
Gaston
On Feb 2, 12:23 am, Mitch <mitc...@gmail.com> wrote:
> A 5 Mhz AM transmitter that can modulate 100Hz and 1000Hz, putting out
> a few milliwatts, would be perfect. I know little about analog
> electronics. Someone else will have to do that part.
>
>
> > If I were you I would spend a little bit of time to work out how a
> > simple WWV transmitter can be made. It really is only a small step to
> > go from the point where you are now. It's a beautiful clock and if you
> > have a GPS to WWV converter (RF), the clock can be used globally and
> > is still doing the thing it was designed for, it's just that it is a
> > local WWV station :-).
> > Michel
> > simple WWV transmitter can be made. It really is only a small step to
> > go from the point where you are now. It's a beautiful clock and if you
> > have a GPS to WWV converter (RF), the clock can be used globally and
> > is still doing the thing it was designed for, it's just that it is a
> > local WWV station :-).
>
Cobra007
Feb 2, 2012, 4:30:02 PM2/2/12
to neonixie-l
That is of course *very* true :-)
Should that be a problem in any country, you could always take the HF
output signal and plug it straight into the antenna input of the
clock.
As far as I know, for the US, an AM transmitter below 15MHz and under
100mW does not need a license. The described transmitter is probably
less than 10mW.
Michel
Should that be a problem in any country, you could always take the HF
output signal and plug it straight into the antenna input of the
clock.
As far as I know, for the US, an AM transmitter below 15MHz and under
100mW does not need a license. The described transmitter is probably
less than 10mW.
Michel
Cobra007
Feb 2, 2012, 7:24:08 PM2/2/12
to neonixie-l
Do you want to trade one of your GC-1000 clocks for the described
transmitter?
Michel
transmitter?
Michel
Mitch
Feb 2, 2012, 8:21:34 PM2/2/12
to neonixie-l
Thanks for the offer. I think I'll keep the Arudino connected directly
to one clock, for the time being. If someone wants to work on a design
for a very low power transmitter, I'll take care of the digital side
and we will end up with something very interesting. A simple
transmitter with a 555 timer that can be selected to oscillate at 100
or 1000hz shouldn't be difficult for some with analog circuit
experience. As Cobra007 mentioned, it is probably legal, but that
should be verified.
The software is working perfectly now. The Time.h library is updating
frames and the clock is tracking perfectly. I don't think I've ever
seen the hi-spec light illuminated before this. The time library is
using the Arduino's internal clock for now until the GPS arrives. The
program still needs some work though, it will be better using
interrupts. Hopefully I'll have time to finish it next week.
to one clock, for the time being. If someone wants to work on a design
for a very low power transmitter, I'll take care of the digital side
and we will end up with something very interesting. A simple
transmitter with a 555 timer that can be selected to oscillate at 100
or 1000hz shouldn't be difficult for some with analog circuit
experience. As Cobra007 mentioned, it is probably legal, but that
should be verified.
The software is working perfectly now. The Time.h library is updating
frames and the clock is tracking perfectly. I don't think I've ever
seen the hi-spec light illuminated before this. The time library is
using the Arduino's internal clock for now until the GPS arrives. The
program still needs some work though, it will be better using
interrupts. Hopefully I'll have time to finish it next week.
Cobra007
Feb 2, 2012, 9:25:02 PM2/2/12
to neonixie-l
It is actually quite an interesting project. I had a look on ebay for
other WWV clocks to do some experimenting but it looks like the
majority uses the WWVB signal with is broadcast-ed at 60kHz and
slightly different from WWV.
Do you know of other clocks that use WWV rather than WWVB?
In Australia we don't have a time broadcasting station, so the only
option is GPS.
Michel
other WWV clocks to do some experimenting but it looks like the
majority uses the WWVB signal with is broadcast-ed at 60kHz and
slightly different from WWV.
Do you know of other clocks that use WWV rather than WWVB?
In Australia we don't have a time broadcasting station, so the only
option is GPS.
Michel
Mitch
Feb 2, 2012, 11:56:43 PM2/2/12
to neonixie-l
I haven't checked, but it should be just as easy to make a WWVB
version.
version.
Cobra007
Feb 3, 2012, 2:58:01 AM2/3/12
to neonixie-l
Well that is true, but I won't be able to check if it will work on
GC-1000 clocks :-)
GC-1000 clocks :-)
Reply all
Reply to author
Forward
0 new messages