Not very reliable ZigBee Locks.

I wonder if ZIM/lock performance is any different with a single lock versus several. I only have one lock and it's about 30 feet from the ZIM (with two walls in between and a ceiling/floor above) and it never misses a transmission.

I hear you on the costs. The price premium of the Zigbee lock versus standard lock is no where close to what it should be for adding a cheap Zigbee module.
 
I don't think the number of locks make much difference, since there is no mesh networking, but its possible Wi-Fi can make a difference, they both use the 2.4GHz.
 
I only have the one lock and the one ZIM - the only two Zigbee devices I have. I do have 2 wifi devices as well as Sonos units (not sure of the frequency Sonos uses).
I hear you regarding the codes too! I was concerned that the instructions explained only how to add 2 codes and it then suggested using the system controller.
I emailed support from the reseller and received the same information. Lucky I only need to have an extra couple of codes.

It would have been nice to have closer integration with the codes to see who unlocked the door in the events.

I think i'm just going to mount the ZIM by the door than risk spending another couple of hundred dollars and getting no better results.
 
I'm starting to think that just buying the regular non-Zigbee lock, hacking into it, and just wiring it to a HAI relay might have been the best plan. :unsure:

Also, some other "features" about this lock I noticed.
1) I wouldn't use 4 digit codes. With only 5 buttons, and if you had 10 codes, that is only a 1 in 62 chance that a random guess would open the door. It does lock out for 60 seconds with three wrong guesses, but even still, that means on average, someone could break in after an average of less than 15 minutes and 30 minutes worse case.
2) Of course that assumes they just don't pick the key lock, which from what I've seen is pretty easy.
3) If that wasn't bad enough, I've noticed on one of mine, that if it is not locked ALL the way, and sometimes it doesn't lock all the way when locked electronically, that its easy to just push the deadbolt back in with little force.

So in my opinion, this lock is just one step higher than leaving your door unlocked, and sticking a note on it saying "Please don't come in." :eek:
 
1) I wouldn't use 4 digit codes. With only 5 buttons, and if you had 10 codes, that is only a 1 in 62 chance that a random guess would open the door. It does lock out for 60 seconds with three wrong guesses, but even still, that means on average, someone could break in after an average of less than 15 minutes and 30 minutes worse case.


Ano,
With a 4 digit code and 5 buttons, there are 625 possibilities. So having 10 codes mean you have 1/615 chance to guess a correct one. I haven't had caffeine yet this morning so hopefully my math is correct.

I am aware the lock does not have a high powered radio, but am a bit surprised that 10 feet is your maximum range. With the 4 locks I have around here I am able to get at least 5 times that when talking to another HAI device. Environment plays a big role in RF strength, but if the samples I have around here are accurate examples of what is in the field then I would not suspect that 10 feet would be the norm.

Your reports however contradict my expectations so I am going to run a few more tests with different environmental effects to see if I can duplicate your findings.

Ryan
 
10 ft. reliably, sometimes 30 ft. but sometimes not. The ZIM is in a room with lots of electrical noise and I haven't had the chance to move it.

OK. this is my calculations on the lock. Yes if you had one 4 digit user code you'd have a 1 in 625 chance of hitting the code any one time. If you have 10 codes you increase the chances by a factor of 10 so its a 1 in 62.5 chance. You can enter 3 codes before your locked out for one minute. So lets say you can enter a code in 10 seconds, so you enter 3 codes in 30 seconds and you get locked out for a minute longer, that's 90 seconds for 3 codes. If you had real bad luck you may have to enter 625 - 10 or 615 codes, however.
 
Ryan, the absolute best distance I've had is around 10 feet away in the same room.

I moved the ZIM near the front door so I could pair and test the lock, which has no metal nor rf to speak of (except a wifi AP and sonos unit around 30 feet away) and if I move the ZIM more than a few feet (10 max) away it no longer works. I originally wanted to place the ZIM in the roof space above the door, which is about 5 foot directly above through a single 10mm plaster sheet - even that was enough to stop the signal.

As I mentioned in an earlier post, I'm aware that that the battery devices have lower powered radios and "sleep" and they supposedly work better with other powered Zigbee devices that can route, so I was contemplating purchasing some plug in LCM's, but I'm not confident given the distance would actually be further (around 15 feet) than I'm having issues with now, that I'd see an improvement.
 
arakasi55 and ano,

If you are using the lock on a metal security door it may be attenuating the RF. The RF daughter card appears to be in the battery box area. Try removing the battery cover to see if range improves. If it does, you may want to consider spacing the lock away from the door. I used 1/8" rubber gasket material to provide an apeture for the RF signal (different lock manufacturer).
 
All my doors are solid wood.

When we say lock to ZIM distance is poor, I wouldn't jump to the conclusion that its the lock that is the problem. The Smartenit devices I have have an LED that flashes when a lock or ZIM goes in setup mode. The Smartenit device can "see" both the ZIM and locks for quite a bit further than 10 feet. I haven't run tests but its 30+ feet. In other words, the Smaertenit device can see a lock in setup mode much further than the ZIM seems to be seeing it, so something else might be at play here. Sometimes its hard to know exactly what is going on.

Just for interest, what Omni FW are you all running? I have v3.10A and I suspect it might have Zigbee problems.

I have a router plug within 5 feet of each lock and its still not working reliably. Sometime the panel can lock or unlock a lock but it can take 10+ seconds to do it. Other times the panel says its locked or unlocked when in fact it has no idea. I'm getting v3.10D firmware to see if that helps at all.
 
I'm running 3.10D.
My door is also solid wood and currently have the battery cover off the lock, although I've found little difference in performance. I suspect this is because the cover is plastic
 
I upgraded my firmware from 3.10A to 3.10D and things are much improved. At the very least, the panel now knows when a Zigbee devices is connected and when it is not. 3.10D also displays the power uses by Zigbee loads, al least Zigbee loads on HAI plugs. (Others are reported as zero watts.) But all is not great, for some reason 3.10D breaks my CQC software from connecting with the panel. I went back to 3.10A until I can get that resolved.
 
Was something about the protocol change in a breaking way in 3.10A? I haven't had a chance to look at it, but that would be unfortunate if it happened.
 
Any idea how to remove an access code from a Kwikset lock? I accidentally entered a 4-digit code, and I've tried to overwrite it several times, but it keeps working. I'm not even certain which code location it's stored in. There is some vague mention in the instructions that codes can be managed "through a network controller", but I can't find any info on what that actually means.

I'm concerned in general about how easy it is to add codes to these locks. It takes about 10 seconds to pop off the cover, hit the program button a few times to advance to a unused code location, and add a new code. The manual says to check regularly for new unauthorized access codes. So I'm supposed to check for new codes every time a visitor leaves my house? Pretty impractical, and perhaps not even possible. There really should be a requirement enter a "master code" first before adding new codes.
 
I had to laugh when I read your post, I'm not laughing at you, I'm laughing at the lack of attention that companies take when they produce new products now-a-days. Its like "get this piece of #@*$% out the door, now lets work on the next."

You didn't even mention the best part, how their stupid "Users Guide" doesn't even bother to tell you how to pair the lock with a controller.

Anyway, the ability to manage the lock "through a network controller" was a figment of someone's imagination. Maybe the "User's Guide" author, likely based in China, thought that this would be a cool feature.

I reread the "Users Guide" and I see the line "Deleting a code can only be done from the system controller" so that would imply you might be screwed. Keep in mind, in theory, this lock can store 30 codes so you may be really screwed.

I would try taking out the batteries, press the programming button a few times, and letting it sit for a bit. I can swear that it looses all the codes when you do that, which is another problem, but lets not split hairs. What do you want for a lock that only costs $250, you expect it to work for that low a price?

The manual says to check regularly for new unauthorized access codes. So I'm supposed to check for new codes every time a visitor leaves my house? Pretty impractical, and perhaps not even possible.

Yeah, maybe they expect that after your cleaned up from the party, you will go outside and test every code. I guess were back to the day-dreaming guy in China again.

Also, you might try something longer than 4 digits. Since it has multiple numbers on each button, that increases the odds someone will guess your code. I'd stick to 5 digit numbers or more.
 
Back
Top