Impinj R1000

[Craig MacIntyre]

I acquired a really clean one to use as a secondary reader.  For some reason I was under the impression that they did POE as well ... alas not!  So any suggestions for a power supply?  The Impinj factory one is about $125 ... was hoping for something a bit more economical if possible.

Sometimes not reading the fine print is detrimental to ones wallet.

[Brandon Daniels]

I just use the standard (factory) Impinj power supply sold on Atlas RFID.  I love how they're interchangeable between all Impinj readers (R420s, R1000s, etc.)  

And they screw-lock into the back of the reader so that the power connector can't slip out or be accidentally tripped. You might find knockoffs on eBay or Amazon.  

But those ones don't seem to have the screw-lock like the standard power supply does.  

And since powering the reader is so crucial, I just feel like I owe to my customers and to my craft to use the most reliable means to power the readers.

So I just bite the bullet and take the hits to my wallet.  Much better than having to deal with an unhappy client :)

[The Road Hunters]

any review about how the impinj reader (r1000/r420 etc) performs with your software?

[Brandon Daniels]

Perfectly!  No complaints...

The R420 is newer than the R1000.  But they both seem equally reliable with Brian's software in my experience.  I've never had any issues with them.

I've always used R420s as my main readers.  And I typically use the R1000s as backups since they're a little older and a more clunky than the newer, lighter R420.

Performance wise they seem to be just as good as the Motorola readers.  However I originally opted for the Impinj readers instead because they seem to be a lot less quirky than the Motorola's 

based upon what I've read in this forum. Bottom line, if you setup the Impinj's correctly and plug everything in correctly, they simply just work.  Whenever I'm timing a race, I simply plug them up and forget about them.

The other thing to take into consideration is that many of the chips used in the inlays/tags we all use are Impinj chips (e.g., M750, R6, etc.)  So obviously the Impinj chips are designed to work pretty 

well with the Impinj readers in general.

There are a few downsides though...

• The Impinj's max out a 4 antenna ports.  You'd have to buy an antenna hub to hook up more antennas to 1 reader or buy more readers to run more than 4 antennas.
• New readers are very expensive.  And sometimes used gear on eBay can be hard to come by.  But otherwise I've found a lot of great deals on eBay. But be careful and make sure there's a good return policy if it doesn't work out.
• For used gear, product support is almost nonexistent especially for the discontinued R1000.  Aside from this forum, a handful of articles on the Impinj website, and a few Youtube videos, you're kinda on your own if you do run into any issues.

[Craig MacIntyre]

The R420 and R220 have been bullet proof.  I find them faster than the Motorola 9500 and 7400 they replaced ... but that could be anecdotal.  They are small, work great POE and I have really had no issues.  As Brandon noted, if you use Impinj chips you actually do have an advantage as the R420 has logic built in that modifies some of the operations to match what you are doing.  Don't claim to understand it but it may be why when using M750 chips it is amazingly accurate and robust.

I have heard that the R700 is even better.

As Brandon noted, limited to 4 ports.  However, one upside of that is you never get sucked into the long finish line 8 port issues due to multiplexing.  I have used 2 readers on two mats and had great results.

[blueridgetiming]

I wonder if you could all elaborate on the types of the event that you are using the R420 for. At this point (for me) using the R420 as a backup, the best that I've had is 90% read rates in XC. To this end, I wouldn't consider the R420 as a primary source reader.

[Brandon Daniels]

The biggest race I had this year was a summer road race with 600-700 people.  Didn't miss a single read.

It's all about the setup though.  At that finish line I had 1 row of 3 mat antennas + a panel side antenna going into the main R420.  Then, behind that I had another row of 4 mats going into a back up R420.  Both readers were connected to my gigabit ethernet switch using shielded CAT-6A cables.  

In fact, I also had 2 more rows of mats and readers located ~50 meters away because they also wanted a chip start.  So readers #3 and #4 were wired into another ethernet switch at the start line.   And I joined the two ethernet switches together with a really long CAT-6A cable.  ART handled all 4 R420 readers perfectly!

In cross country, I typically used the "1 row of 3 mat antennas + a panel side antenna" into one R420 setup method.  In small to medium sized races I typically got 98-100% read rates.  But after thorough investigations, I could always blame the few misses on poor antenna placement, loose antenna cables or mats being wet, but never the reader itself.  Anytime there was a race with missed reads, I could always make some kind of adjustment to get 100% the next race.

In fact early on, when I timed summer meets, I also learned that the readers don't work well if they get too hot.  So I purchased a few high powered laptop cooling fans from Amazon that the readers sit on top of in their timing boxes.  Haven't had a problem with the readers overheating and missing a lot of reads ever since.

And I also learned that if I wanted 100% consistently (e.g., for big races and for state qualifying races) that adding that second row of 4 mat antennas can make a huge difference in tough scenarios, especially in crowded races where runners come through the finish line very fast.  The timing company that times our state meet also times the NCAA DI CC Championships, and they always use 2 rows of RR mats, even for their all of their split locations.  Using 2 rows of mats has been practically bulletproof for me.  So far I'd say it's 99.9+% effective.

[The Road Hunters]

thank you guys, really appreciate it. 
just got my second hand impinj r1000, but still looking for power reader and i Hope it will work great as you mentioned.
I wonder, have you ever set up 1 long finish line (8m or more) into 2 impinj r420 with ART?

[Brandon Daniels]

Haven't need to do 1 long finish line yet.  But it's very easy.  All you need is a good ethernet switch (e.g., linksys, d-link, cisco, etc.)

Then just run power and ethernet under the mats to the far reader.

I've done 4 rows of mats into of mats into 4 R420s (2 rows for chip start + 2 rows for finish), all connected into 2 ethernet switches.  It works perfectly fine in ART.

I use shielded CAT-6A for ethernet cable because it has a very thick outer jacket.  And it has thicker copper conductors inside.  So it can take a lot of outdoor abuse at road races.

[The Road Hunters]

thank you, I've done 2 rows also with ethernet switch also.

my case is I will time a race with 14 meters gate.
8.000 runners already registered, questions are :
1. what is the best set up for this?
2. what if, I use 14 meters feibot matt at the one line and connecting it into 2 readers (fx9500). will it cause  problem?
thank you

[Brandon Daniels]

Haha I've never done a wide 14 meter finish line or 8,000+ runners.

In my opinion, it would totally depend on what kind of read rate you're wanting to achieve.  With 8000 runners, a 98% read rate still means that you miss 160 runners.  Is that acceptable to you?

Personally I wouldn't trust just 2 readers to do THAT wide of a finish line.  I'm a very cautious/conservative timer, so I'd want a minimum of 2 rows in that scenario.  And if I'm trying to be bulletproof, or as close to perfect as possible, (call me crazy, but) I'd I want three rows of mats to cover that much finish line for that many people.

Others who have the FX9500 may tell me I'm wrong and that your idea should work flawlessly.  So maybe I'm wrong, but I just don't see that working very well for 14 meters wide & 8,000 ppl.  But either way you'll definitely want to have a pretty good back up system and some video cameras recording the finish line.  

[The Road Hunters]

Unfortunately, I dont have backup for this.

Usually the gate is just 6 meters.
Hope everything will be ok.

[Brandon Daniels]

If it was me, then I'd just tell them that I have a total of 14 meters of mats and that the widest I could go is 7 meters wide.  Because, in order to make this reasonable, I'd need to have a second row of 7 meters of mats behind the first row of 7 meters to get as close to a 100% read rate as possible.

If they said that 7 meters wide is too small, then I'd offer to do 9 meters for the first row and then funnel them into a second row of mats that's 5 meters wide (or 8m & 6m).   But to trust just one row of mats to time 8,000 people and to expect everything to go right (without a backup)? I don't know lol 

[Craig MacIntyre]

I am no expert but do know a lot of what not to do!

First, accuracy means many things ... if you are meaning highest read rate then Brandon is spot on.  Brian designed his system around NOT using bibs specifically because they are expensive and give the worst read rates of just about any method.  Unfortunately that seems to be the way RD's have taken the market.  In our experience, with bibs, the absolute best read rates come from overhead antennae's - you can angle them enough to give fairly good reads on decently placed bibs.

If by accuracy you mean getting correct finish times then passive chips are a terrible solution.  You need active chips and/or high speed video.

The good thing about chip timing a big event is that the start chute is SLOW after the first few rows get through.  As Brandon noted you need multiple rows and some backup.  Feibot mats bought from Brians deal are inferior to RaceResults because they have gaps.  On the other hand they are more efficient and when used with multiple rows (offset) and side antennaes they work great.  

I don't know what you are being paid but I would consider highering extra crew - maybe in the form of another timing company to ensure you have enough people, readers, equipment.  If you tank a 70 person 5km you have reasonable backup options, etc.  It won't kill your business.  If you blow this one it will.  On the flip side, if you succeed at this one it could vault you to the next level!  Congrats ... I dream of being offered a bigger event (maybe not of that magnitude!)  LOL 

[Graham Hall]

Ten years ago, I had 8 x1m mats so I set up a 8m wide start line (because I could!) for about 1200 runners - it was like cavalry charge, with possibility for trips and falls clear to be seen, and quite a few missed reads

 The following year, using the same mats I set up a double line of 4m and funnelled the same sized field over the narrower start. The result was a more orderly start, with runners able to find space to settle into their pace almost immediately, and less potential for trips and falls – plus far fewer missed reads.

 Since then I have noticed many of the really big city events have now adopted similar narrow start lines.

 The old 1m mats have gone and I now use 2 x 4m Feibot mats each with 8 antennas connected to an FX9600 and I get read rates of 99% +.

 So, in this context I cannot conceive the need for a finish line that is any wider!

[Jim Hoppe]

Brandon,

I'm very impressed by your read rates. I have some questions about using the multiple rows of mat antennas. 

How far apart are you spacing the mats and antennas?

If you miss on the first row(s) would the later reads be considered late and inaccurate? I've heard that a late read is better than no read!

Do you get crosstalk and interference as Bryan has discussed in some of his videos.

Are you using only single chips or multiples, on bibs or shoe/hips chips?

Thanks

Jim

[Brandon Daniels]

Hey Jim,

Here's a setup pic I took after a race I timed a few weeks ago...

Those mats are maybe 10-12 feet apart or so.  I usually just pick a spot beyond the first row simply based upon my intuition.  That has always worked for me!

Late reads haven't been an issue. Fortunately the first row gets 98-100% of finishers in most races.  The >2% that the first row might miss usually aren't wearing their bib correctly or simply wouldn't care about a late read anyway. 

Remember, many of us are timing fun runs and charity events where no one's gonna know or care about their time maybe being a 1/2 second slower.

If the race was that serious, the race director wouldn't/shouldn't be relying on chip timing for accuracy in the first place.  

They'd use a line scan camera (like the purple Finish Lynx camera on the tripod in my pic) which is way more accurate than any chip system by far (0.001).

The Finish Lynx camera is overkill for these fun runs where no one's gonna complain about their time anyway.  But I always like to set it up as a backup, just in case. Plus it's my favorite toy :)

In high school cross country, I always decide the official results based upon the Finish Lynx photo finish anyway, not the chip reads.  And it's always there as a backup for my serious road races in case there's any question about chip reads.

I haven't gotten any cross talk or interference.  I space out the mats and side antennas far enough apart to where they don't interfere with each other.  Plus, the Impinj readers are designed to automatically sense when there are other readers nearby (e.g., two R420s).  And when the readers sense each other, they automatically use completely different frequencies to help prevent interference.  

Not only that, I alternate polarities when I setup multiple side antennas.  Some of the side antennas are LHCP (left hand circular polarity) and some are RHCP (right hand circular polarity).  That further prevents any interference.  For antenna cabling I use LMR400-UF which is pretty low loss, thick/sturdy and good against interference. For ethernet, I use shielded CAT-6A which is also pretty good against electromagnetic interference.  I spare no expense on sturdy cabling because it gives me peace of mind.

For chips I use two vertical chips on the front of the bib (about 5 to 6 inches apart).

[The Road Hunters]

thank you for sharing.
we have finished timing and get the worse result so far.
because of the gate is 12 meters wide.
here is the photo of the finish line :


I use 2 side antenas, 4 feibot matts, and 6 times-7 matts, and connected to two fx9500 readers.
Multiplexing is really big problem, since we have to cover 12 meters start and finish line.
luckyly, I backed it up with 3 photographers and 2 videographers.

I heard from my friend who used impinj r420, it can be programmed to avoid multiplexing by programming into reader CAP.
CMIWW

[Brandon Daniels]

Yeah, that's why I urged you to use 2-3 rows of mats in a scenario like this.  I never thought this would've actually worked with just 1 row like this no matter what.

And even from the first conversation with the client, had they told me there would be thousands of participants in the race, the first question out of 

my mouth would've been, "How wide will the finish line be?"  So I would've learned that early on and have prepared well in advanced to make sure I had enough equipment.

As Craig suggested, you really should've reached out to another timing company for help or to have rented enough equipment to add another row.

I'd rather lose money the first year of doing this race than to make a bunch of money this first time but potentially risk losing the repeat business from this client.

But it looks like you've learned a lot by going through this.  Hope you can invest in more equipment, and good luck in the future!

[The Road Hunters]

from this discussion, I come up by buying impinj r1000 reader secondhand. the problem is I can not connect it to my computer. do you have any idea how to reset impinj r1000?
Bcs there is no power button and it seems my secondhand reader has been set to static IP. thank you.

[Tanvir Ahmed]

You can connect to impinj reader with host name  http://speedwayr-xx-yy-zz   where xx-yy-zz is the last three sections of the Speedway reader's MAC address. the default password is impinj, if you can get to the interface you can set the ip address information, mac address should be on the back or side of the reader 

Regards

Tanvir 

[Raphael Leal]

Got the same issue.

I can´t access the interface to change/set the IP address.

The default user/pass is root/impinj but "impinj" is not working.

[Brandon Daniels]

Did you actually perform the factory reset?

https://support.impinj.com/hc/en-us/articles/202756458-Performing-a-Factory-Reset-on-Impinj-RAIN-RFID-Readers-and-Gateways-VIDEO

If you successfully perform the factory reset, root/impinj is the username/pass

The factory reset process is tricky so you might have to try again.