EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: croyleje on December 30, 2014, 01:58:17 am
-
hello everyone,
i have finally completed my GPS disciplined 10MHz frequency standard and it is working great but i am not happy with having to have my GPS antenna run across my shop and out onto my deck. i have been looking for ways to receive GPS indoors or maybe make some kind of relay to wireless send the GPS signal from my antenna to my freq standard any ideas or thoughts of how i could get a good GPS signal indoors? i have been reading online that new receivers are better indoors and there is some new Apple tech coming out used to relay GPS signal indoors for use in malls so people can be directed to stores but couldn't find data on how it works. any ideas of how i could do this would be nice at least it would make for a nice intellectual problem. my leading idea is receiving the signal with my Navman Jupiter T Tu60 then sending the signals via bluetooth to my freq standard but i think thats going to effect the timing enough to make the standard useless? dont have any ideas on how i could send the 1PPS and 10KHz signals without losing the accuracy of the GPS lock. well any thoughts or ideas, just thought it might make for an interesting discussion.
Jason
-
If you want accurate measurements you will need the external antenna, however if you are near a window you will probably pick up some multipath signal. Most of the better industrial receivers try to implement some sort of multipath rejection scheme to reduce errors, which the consumer grade receivers you mention are probably not concerned with.
I don't see how your Bluetooth solution could work.
EDIT:
On second thought I seem to recall GPS repeaters or 're-radiators' were in place at one of my old work places. Perhaps google those terms. You would still need the GPS antenna which is connected to the re-radiator which is pointed at your receiver, the re-radiator would need line of sight, but otherwise works at a distance. Don't think its cheap though.
-
thanks RROBOT i didnt even think of a repeater with a quick search i found several online and the price is very reasonable. i will hook one up on my deck and mount the other "re-radiator" in my shop should work great thanks sometimes the answer is simpler then you think thanks
jason
-
I seem to recall the re-radiator in the office I worked was physically connected to a antenna on the roof via coax. Inside the building the re-radiator rebroadcast the signal. The one I am familiar with could not receive signals from the antenna wirelessly. You may want to verify with your supplier, if thats what you need. This was a long time ago and I do not know much about them.
-
If your deck changes location I would get out of the building quick :)
Seriously now, if you only need to broadcast your location and it's a fixed location you don't need all that. Now if the customer wants to find you within a large building then the re-radiator will make sense but only for the client's GPS receivers.
-
It sounds like he wants the time signal, not location.
AFAIK most repeater systems will give the location of the receiving antenna. GPS location is based on very precisely comparing time-of-flight from multiple satellites - if the signal isn't coming straight to you, it doesn't work at all. At least not at any accuracy.
-
Oops, sorry I should have paid more attention to the "disciplined 10MHz frequency standard".
As for the inaccuracy if you only use one repeater I agree with you, it will place you at a lower altitude than the antenna. If various repeaters are around the building then you will have the same satellite signal with different timings so not sure how that will work.
But for getting the time signal it should be fine, I would like to see what results the OP gets and how is tested for accuracy.
Edit: looking at the OPs avatar I guess it will involve a Rubidium frequency standard ;)
-
You will need a GPS repeater. They run about $1500. You will have to make sure that the repeater does not broadcast outside, or you will have created a GPS jammer. An antenna goes onto the roof. Another antenna plus an amp and electronics is connected to it on the inside. A short run of coax connects the two as previously noted. It will replicate the location signal of the antenna on the roof indoors. I don't know how well this works for time sync. It is typically used so that vehicles parked indoors can leave the building with their GPS locked-up.
For your application, it would be far simpler to locate just the antenna outdoors and run coax to your receiver. That's how it's done everywhere. Go look at any cell site. You will see a little white dome or cone attached to a steel post outside. That's the NTP antenna. They are typically powered with something like 5V on the center conductor so that the active antenna can boost the signal to compensate for the line loss. These antennas are common and fairly cheap as I recall.
We all stick GPS antennas in the window as a lazy way to do development. I'm as guilty as anyone. For an NTP server, that won't be a reliable, long term solution. The antenna simply needs a clear view of the sky. No bolting it right against the side of a building.
If you're in a lightning prone area, you will need to ground the shield and install a lightning arrestor to protect people as well as your gear inside. That equipment should come as part of the repeater kit.
Here's an example of a NTP antenna that's powered:
http://store.novotech.com/collections/trimble/products/bullet-antenna-5-vdc-f (http://store.novotech.com/collections/trimble/products/bullet-antenna-5-vdc-f)
-
The GPS modules with helical antennas tend to work better indoors.
-
Whoops...I just learned something new. Now you're going to need an FCC experimental license to operate a GPS repeater.
http://www.gpssource.com/faqs/7 (http://www.gpssource.com/faqs/7)
GPS Source in the link above pretty much is the leader in repeaters, so if you decide to go that route, they would a good place to start.
-
Do you not have any way to mount an antenna outside? That is probably the cheapest and most reliable option. My antenna was about $50 from eBay and I ran a long length of coax into my basement lab (I have an access tube from the attic to the basement ...) and connected it to my GPSDO.
Anthony
-
Why not just mount the receiver etc. in a good place, and distribute the 10MHz signal via cable?
-
Do you need 10^-13 accuracy all the time? I use mine as a reference once a year to calibrate an OXCO in my frequency counter; the OXCO is always on anyway.
-
I just read and re-read the OP...
It seems to me that he is approaching the problem with the wrong technology (GPS) to implement location/tracking within an enclosed/covered environment.
The frequencies and very low RX signal levels involved are entirely incompatible with the expected product goals .
Look at lower spec 433M / 2.4G transponders to give quite accurate 'geo'location withn the facility - using RSSI and triangulation techniques.
A colleague of mine did something very similar - a couple of years ago with low-power/long-life RFID tags, and we could get approx 1m (3') resolution in an office structure using four fixed (wired) transponders and multiple moving targets.
But even then due to the wall materials and other variables - it was necessary to initially calibrate - and tune the RX environment.
-
The OP is using GPS to calibrate a frequency standard. Location services are not required.
The best, first way to do this is with a powered antenna with a piece of coax to the receiver. I know an antenna isn't the most attractive roof ornament, but the Trimble antenna will mount directly onto a threaded pipe so that the mount will act as the conduit if that makes anything better.
-
Outdoor antennas work a lot better and if they are well amplified then very thin coax can be used to connect them.
The advantage of GPS antennas is that they need a good view of the sky but this does allow them to be hidden in things like chimney pots and they are very small.
The OPs main problem seems to be that of trailing the cable. One possible solution is to locate the GPSDO close to the antenna and only use it to calibrate an OCXO periodically which can be then located indoors.
One issue I found with doing a neat job is getting cables from outdoors to indoors - in my case a fitted a type N in an external box hidden under a step and took a very thin co-ax through the door frame to a small sma type connector indoors and only connect a cable when I want to use it. The cable from my external antenna is then mainly hidden and taken under a path before connecting to the type N connector. Using the connectors in this way allows the use of a low loss cable for most of the external run and using the thin (high loss) cable only where needed.
-
What type of roof is over your shop? I get a pretty good signal with a timing antenna just putting it high up near the ceiling, in a wooden house.
-
ok ill try to answer the questions asked. my shop is simple a room in my basement to make my freq standard work right now i have the antenna run from my shop up through the floor and have the antenna mounted on my deck. with the re-radiator which from what info i can find does rebroadcast the antennas location it would be nice to get away from that long antenna wire and mount the re-radiator on my roof and the internal antenna in my shop giving me gps in my whole shop and making my freq standard basically wireless i could just simply mount the frequency standard antenna on top of the freq standard.
the big question is and the thing i cant seem to get a good answer on is does the "re-radiating" of the GPS signal effect the timing accuracy?
if anyone knows that info would be great. the re-radiators i found on line are not that expensive but don't want to buy one if its going to effect the accuracy.
thank you,
Jason
-
If you are locking to a low frequency signal, 1pps or 10kpps, you could modulate an ISM band transmitter appropriately.
For 10MHz, possibly infrared might work, but some care would be needed at the receiver, and a suitable coding/modulation scheme would be needed.
Any of these methods, though, are going to introduce jitter to your previously pristine standard.
Or buy a rubidium standard instead.
-
ok ill try to answer the questions asked. my shop is simple a room in my basement to make my freq standard work right now i have the antenna run from my shop up through the floor and have the antenna mounted on my deck. with the re-radiator which from what info i can find does rebroadcast the antennas location it would be nice to get away from that long antenna wire and mount the re-radiator on my roof and the internal antenna in my shop giving me gps in my whole shop and making my freq standard basically wireless i could just simply mount the frequency standard antenna on top of the freq standard.
the big question is and the thing i cant seem to get a good answer on is does the "re-radiating" of the GPS signal effect the timing accuracy?
if anyone knows that info would be great. the re-radiators i found on line are not that expensive but don't want to buy one if its going to effect the accuracy.
thank you,
Jason
it depends on whether it reconstructs the signal or acts as a dumb repeater/transponder. If it's the latter, then the timing accuracy isn't affected.
-
forgot one other thing this is on a side note actually but does anyone know of an easy way to generate say a 10KHz freq from the 1PPS signal from the GPS receiver? i was thinking about also building a simple frequency distribution amp with a few commonly used frequencies like 10KHz, 100KHz, 1MHz etc. and use the accurate 1PPS signal to do this rather then dividing down the 10MHz and keeping my rubidium on all the time. just a thought.
Jason
-
here are a few links to what appears to be the most common of the two styles on the market. the more expensive versions appears to also amp the signal but neither gives info on what and how it is rebroadcast i guess i could buy the cheap on and try it but would prefer not to do that. well thanks for all the input and if anyone knows anything about gps "repeaters" or "re-radiators" and what effect they have on timing accuracy or links to that kind of info would be great.
http://www.ebay.com/itm/NEW-GPS-Antenna-Receiver-Repeater-/231344447792?pt=GPS_Antennas&hash=item35dd342130 (http://www.ebay.com/itm/NEW-GPS-Antenna-Receiver-Repeater-/231344447792?pt=GPS_Antennas&hash=item35dd342130)
http://www.ebay.com/itm/GPS-Signal-Repeater-Amplifier-Transfer-Full-Kit-Distance-15-Meter-/131079096239?pt=LH_DefaultDomain_0&hash=item1e84ec47af (http://www.ebay.com/itm/GPS-Signal-Repeater-Amplifier-Transfer-Full-Kit-Distance-15-Meter-/131079096239?pt=LH_DefaultDomain_0&hash=item1e84ec47af)
thanks again,
Jason
-
if anyone knows that info would be great. the re-radiators i found on line are not that expensive but don't want to buy one if its going to effect the accuracy.
thank you,
Jason
it depends on whether it reconstructs the signal or acts as a dumb repeater/transponder. If it's the latter, then the timing accuracy isn't affected.
As you say, if it just introduces a fixed delay then this will be no different from a length of cable. It affects the absolute timing accuracy (some receivers allow you to account for cable delay) but for 10MHz frequency locking it shouldn't matter.
What may matter is if you get odd reflections so that the signal path changes but provided you don't walk in front of the transmitting antenna it should be ok. But you would be putting the transmitting antenna (of the repeater) in a box. If the sides of your room are around 3 m then by my calculation reflections will be delayed by 10 nanosecs per side so if there is a sudden change of path it will introduce a lot of jitter.
Personally I'd still use an active antenna with a thin cable, the cable can be as thin as 2mm as seen in the attached photo. You could drill a very small hole through the floor in a corner where it wouldn't show (much).
EDIT: the other thing I'd have against the wireless approach is that you're introducing extra electrical noise into your electronics shop, personally I try to remove as many sources of noise as possible e.g. by using linear rather than switching supplies where possible and filtering the mains to my workbench.
-
Personally I'd still use an active antenna with a thin cable, the cable can be as thin as 2mm as seen in the attached photo. You could drill a very small hole through the floor in a corner where it wouldn't show (much).
While I realise the OP is averse to this, certainly this is my favourite course of action, assuming it's an active antenna and there's a bias T in the receiver that can power the antenna mounted preamp. If it's a passive antenna, all bets are off the losses at L band in thin coax like RG174 are going to render the signal useless much beyond a few feet.
Or Mike's suggestion, distribute the 10MHz over thin coax, where losses will be insignificant.