Using a USB device 50 to 100 feet from PC?

There seem to be products called “USB power adapters” that put power on the +5 and ground lines, but don’t let you connect any data. They seem to be aimed at products that tried to take advantage of power available on the USB, and then need to be operated when there’s no computer handy. So they take away the communications ability of the bus, turning it into a wall wart.

I can imagine a product that adds its line-derived +5 to the USB and passes data and ground through from the computer. This is what I need (or will make). But I haven’t found such a thing on the Web. I wonder if they exist and have a name?

Let’s talk about power first.
I’m not sure how picky the GPS puck is, but I have seen other devices such as routers run just fine on “one wire” long runs of Cat5e, provided that the source voltage is correct. You most likely have a 5 pin mini USB type B female connector on the GPS puck, correct? If so, the instructions provided by Beowulff show which pins (1 and 5) are 5V DC+ and ground over USB. There are lots of other devices, such as cell phones, MP3 players, etc. that use a similar arrangement, so 120V AC adapter chargers with USB cables for those devices may work to help you troubleshoot-safe as long as you check the voltage first to make sure.

USB ports are not all created equal, and there is some “negotiation” about power consumption between devices when initially hooked up. Power problems may come up when “daisy chaining” numerous devices, or connections from a keyboard, monitor, non-powered USB hub, etc. But I have seen it be a problem on front-of-case USB connections as well. I know earlier on you mentioned that you are having problems with the USB ports you’re using on your computer. Do you have another machine or known good powered USB hub you could use to troubleshoot it? Does the GPS puck have any lights to indicate it is getting power/charging, and do they light up when connected to the length of Cat5e?

As far as not receiving the signal, I have done this GPS post-correction work in the past, only with Trimble Pathfinder Office and GPSCorrect/GPS Analyst which are very expensive. What kind of software are you using? I’m not sure how the data are actually being transmitted as I’m more familiar with an RS232 or Bluetooth virtual serial port connection, so I may be way off here if the data are being sent as “packets” or some other thing related to USB communication protocol.

But if it is similar to the serial port connection, what baud rate is the data being sent at, and can you try a different setting? I’m not familiar with RINEX files, so I can’t be sure that it is not a different data transmission protocol. Is the GPS puck sending an NMEA0183 string that is being then compiled into the RINEX file, or if not can you set the GPS puck to send a NMEA 0183 string instead? There are some PC (and Pocket PC) applications that can be used to debug the connection. In the past I have used Xport to send the incoming COM port data to multiple applications and the GPS debug feature of ESRI ArcPad to decode the NMEA0183 string, but there are many freeware GPS tools available as well.

Before you go to a lot of effort, you might want to hack up a USB cable and actually measure the voltage at the far end - it may actually be OK, and there’s some timing problem instead. If you find that the voltage is too low, you can just test it with a 9v battery attached to a 5v regulator (LM7805 or equivalent, from Radio Shack), or maybe even use a battery pack made up of 4 NiMH batteries.

Stan and beowulff, great posts, thanks.

I did find a USB 1.1 hub in the basement, which has a plug in 7.5 V power supply. I added it to the setup and now it works.

That is, my setup contains

  1. Desktop PC with 1.1 USB ports
  2. One end of the USB to Cat5 adapter kit
  3. 50’ Geek Squad premade Cat6 cable (I assume 6 means somewhat better bandwidth, dropin compatible with 5e)
  4. Other end of the USB to Cat5 adapter kit
  5. USB cable, one end flat to fit USB ports on a PC, the other end square with 2 flatted corners, like it would fit the USB port on a printer
  6. USB 1.1 hub with #5 plugged into its source port and its wall wart on an extension cord plugged into 120 VAC
  7. GPS receiver’s dedicated USB cord plugged into one of the general purpose ports on the hub
  8. GPS receiver puck (the connection between the puck and its cord doesn’t look familiar to me, maybe it’s the mini 5 wire USB you mentioned? I’m logging data now and don’t want to break it to look)

I think the fact that this works is consistent with the idea that there was too much voltage drop in the line to power the puck, but doesn’t prove that idea. For example, I guess there could be a timing problem, and the hub is in some way more robust than the puck in this regard, so it doesn’t have the same timing problem. Anybody want to weigh in on this? The right way to answer it is to cut open a USB cable and measure the voltage there as beowulff suggests.

I guess this is obvious, but AFAIK the use of the Cat5e/6 cable is proprietary inside the adapter kit and we can’t actually assume anything about what is going on in that cable. I mean, it wouldn’t be Ethernet traffic.

I think I will want to create a more permanent installation with a longer-than-50 cable I will make up. So, I think I do need to check the voltage, before I go fishing cables that won’t work.

Stan Doubt, I think this puck is communicating proprietary data to the program that talks to it. It’s GPS data, but it’s not NMEA. The GPS data includes pseudoranges and carrier phase data and Doppler shift data, as well as Almanac data, all of which aren’t part of NMEA informatio (well, the 183 version that I know about, anyway). The products are an EarthMate receiver and GPS PostPro 2.0 software from DeLorme. They say they are transferring and logging “RAW” data into a “.raw” file format which isn’t useable by other company’s products. I just interpret this to mean it’s their proprietary format, undocumented to the end user and reserving the right to mess with future version changes.

I don’t know enough about USB to even know if it has baud, handshaking, parity, and other features analogous to RS232. Certainly, not anything they describe in the product docs.

RINEX files are text files that have been formatted in certain ways. The receiver probably isn’t sending anything that goes directly into the RINEX file without being processed in some way first. The software, while it is logging, also displays the “raw” data, and it is displayed as an unbroken stream of numerals and A-F letter characters, no spaces, wrapping to fit the window. I presume these are hex digits but they don’t say. There are many sequential zeros in big chunks of the data, especially when I first turn it on.

Left out a couple details.

RINEX is “Receiver INdependent EXchange format”. It’s ASCII text. I pasted the first few lines of a file I saved earlier into the bottom of this post. There is a block that is sort of repeated twice at the end; the rest of the file has several thousand more blocks like this. Their length varies. I think there is a line for each satellite that is being included in the dataset, 3 variables for each satellite. I think the different line that starts each block shows the date, time down to the hundreds of nanoseconds, and don’t know what else.

I think I can set the puck to send NMEA instead. I use lots of NMEA from handheld Garmin units and mast-mounted Garmin GPS17 units that can also send binary data and 1 PPS. I use RealTerm, which is a profoundly versatile and powerful freeware comm program that, refreshingly, does not assume it’s dialing up a BBS. It can send characters or binary data in time-specific ways and do all sorts of other special things. I highly recommend it - though there are a few things in that that confuse me too, I’m ashamed to admit. Anyway, this certaily represents another avenue of approach if things don’t clear up.

Thanks!!!
2.10 OBSERVATION DATA G (GPS) RINEX VERSION / TYPE
DeLorme GPS V3.1 DeLorme 27-Jan-08 00:22:01 PGM / RUN BY / DATE
xxxx MARKER NAME
xxxx MARKER NUMBER
PostPro DeLorme OBSERVER / AGENCY
1234 Earthmate 0.0 REC # / TYPE / VERS
xxxx xxxx ANT # / TYPE
1185240.0790 -4774799.1762 4045593.4299 APPROX POSITION XYZ
0.0000 0.0000 0.0000 ANTENNA: DELTA H/E/N
1 0 WAVELENGTH FACT L1/2
3 C1 L1 D1 # / TYPES OF OBSERV
2 INTERVAL
END OF HEADER
08 1 26 21 41 23.96974 0 4G15G24G26G29
50994088.613 6859170.784 956.379
50997845.087 39972137.586 460.220
50367568.699 13028891.617 427.519
52172575.248 11486025.381 -1867.741
08 1 26 21 41 25.9697381 0 4G15G24G26G29
51030546.934 7050746.184 954.860
51034491.940 40164704.681 459.060
50404220.892 13221525.189 425.641
52210103.232 11683248.698 -1869.126

Pick an antenna spot. Have a survey company come out and sit on it while tired into the nearest HARN marker or two and tell them the accuracy you want to get to. 1/10 inch or less, no problem if they can do it on off time or whatever you can work out with them. Then you have a really ‘tight’ ‘home’ plate to use for all the other stuff and you can compare your results to your own observations and get a better handle on how long you need to sit on other places to get the accuracy you want.

Might be more cost effective too…

spellinc

YMMV

What about this Network USB hub? It looks like it should be able to connect USB devices to anywhere on a network.

>nearest HARN marker

What’s that? How do I find them? Since surveying is my hobby, I might want to do this myself, if I can.

ticker, I like that. I have my network going several places it might be nice to monitor, either continuously or for an experiment. I have seen network USB print servers, which are much cheaper - you think they are different, in some way that limits their application to printers?

One place to learn a bit: http://www.co.hamilton.in.us/services.asp?id=2300&entity=2200

Just stick ( USGU survey HARN monuments ) into google and read till your eyes bleed…

Not sure what GPS puck you are using, but my Holux device is serial, but came with a in-line USB converter. If yours is similar it should be no big thing to extend the serial part of the connection rather than the USB side. IIRC the GPS standard is only 4800 baud, so it long cables shouldn’t cause a problem.

>If yours is similar it should be no big thing to extend the serial part of the connection rather than the USB side.
It’s an Earthmate from DeLorme, USB only.

>IIRC the GPS standard is only 4800 baud, so it long cables shouldn’t cause a problem.
I think NMEA 183 has a standard of 4800 8N1, but this setup can’t use NMEA sentences because the information content is very different. It’s the block of text I posted several items up. NMEA doesn’t deal with any of that.