Hi everyone, I’m using the Sparkfun ublox F9P boards for agricultural equipment guidance. I have a base mounted to my house with a top106 antenna, that has been there since last spring. It works great.
In the past week, I have had two separate Top106 antennas mounted on the top of my tractor fail in under 6 hours of use. The first one I considered a fluke, but the second one is making me suspicious.
There are no signs of physical damage or water ingress. I have tested both antennas on two ublox F9P receivers, with a direct TNC to SMA adapter, to rule out cable problems. They are not capable of detecting any satellites, even in the same location where know good antennas get 12 sats and RTK fix.
I am wondering if something inside the antenna is vibrating loose? Would it be possible for me to open the dome up and resolder, then maybe dip everything in epoxy to prevent a recurrence?
I am not currently in a jam because I have spare ublox L1/L2 antennas, but these big ones are pretty expensive so I’d like to recover them if at all possible.
Anyways, let me know what you guys think. I’ve owned one of them for about a year and the other one for about 6 months. Not sure what the RMA policy on that might be but obviously if you guys want them back I’m willing to send them.
Can also send pictures & videos or upload to YouTube if you want.
This was my issue. To test, you can take a tiny sliver of plastic and slide it between the outside of the pin and the plastic insulator to make the opening tighter. It’s not a permanent solution of course but it’ll at least let you know if the antenna and LNA inside are OK.
If you want to take it apart, you’ll need a dremel and a new enclosure. Luckily though, the assembly is held in the case with 4 screws and even luckier still, the assembly is attached to the TNC connector with a standard MCX connector.
I haven’t got mine re-assembled yet but a standard 6x6x4 PVC junction box should work fine as a new enclosure. I’m going to mount the assembly on the underside of the lid with some standoffs to keep the antenna face itself slightly away from the PVC. A quick test shows that signal strength difference between the original case and the box’x PVC lid is undetectable.
You can then use a bulkhead connector of your choice and adapters to connect the assembly. If the assembly’s antenna reference point is important to you, it’s the bottom of the original case. You can get the measurements from the datasheet and transfer them to your new enclosure.
A few thoughts from inside SparkFun: We’ve sold many TOP106 antennas and have gotten a few defective units back due to water ingress. It’s a thing and why we don’t recommend the TOP106 anymore. Please see the new tri-band kid on the block: https://www.sparkfun.com/products/21801 However, the rate of failure due to water ingress is below 1%. I’ve got one antenna that was returned that has water inside, I’ve also got two units that have been in operation for 3+ years with no issue. YMMV.
I really like etj0’s comments. It’s harsh out there and I would not be surprised if the TNC center pin could become a point of failure. I’ve opened up a few TOP106’s and they are hard to open. So if you do decide to do fault analysis, please consider it to be fairly destructive. I suspect with enough silicon caulk, it could be re-sealed but it would be tricky. I like the idea of a PVC junction box. As long as the antenna elements stay dry, a new plastic enclosure is a fine idea.
Inside the antenna it gets tricky to fix. The MMCX connector could have come unplugged from the TNC bulkhead connector (shown in etj0’s photo) but I’ve never seen that and it’s a pretty tight fit. The MMCX cable is then soldered to the antenna PCB but it’s under a metal RF Shield which is pretty evil to de-solder/remove. All that said, we had to remove the RF shield to take the internal product photos (https://www.sparkfun.com/products/21801), it’s just work. If you plan to replace the MMCX cable, I recommend the SMA to U.FL (https://www.sparkfun.com/products/9145). It’s thinner than the MMCX (which we don’t stock), you’ll cut the U.FL end off, strip and solder to the antenna PCB. But should give you a direct shot if you’re using a SMA extension cable (no more need for the TNC to SMA short cable).
Other points of failure: cables. You’ve already addressed this (nice!) so I don’t think that’s your problem. Just FYI - we started carrying the reinforced cables. These have injection molded boots. Short (https://www.sparkfun.com/products/21739) and long (https://www.sparkfun.com/products/21740). If you’re doing a permanent install, I use the long cable (https://www.sparkfun.com/products/21740) will allow you to avoid one SMA to SMA connection (potential point of failure).
I redeployed mine yesterday inside that 6x6x4 box and it’s performing great. Of course I forgot to take pics first. I used a type N bulkhead connector and MCX to SMA + SMA to N adapters inside because i have tons of them hanging around.
One thing to also remember… If you have a long cable run you’re going to get some expansion/contraction of the center conductor which can cause the connector center pin to lose contact. If that cable run is also vertical, gravity can also play a part.
@sparky Is the SPK6618H going to be calibrated and submitted to ANTEX?
Yes, SPK6618H is currently being calibrated by NOAA’s National Geodetic Survey - using their fancy new robotic arm setup. I’m sure there will be a blog post about it soon.
I used a type N bulkhead connector and MCX to SMA + SMA to N adapters inside because i have tons of them hanging around.
