Warehouse Omni Deployment with Barcode Scanners

[littleyoda]

I have a 618,000 sqft warehouse I need to upgrade wireless to. I'm limited with deploying 9120axi for this one. The survey shows moderate coverage for a -10dB barcode offset with 24 Motorola APs, which are currently at maximum transmit power. APs are currently mounted at the i-beam 32 feet high. No racks in this one. Just mechanical assembly lines.

My design requires to install these at the same height. An ideal design shows I can get away with 30 APs at a 12.5mW transmit. But when you add the barcode offset (-10dB 2x2 802.11ac compatible), the requirement (-67) 5Ghz misses by a lot. This causes me to deploy 20 more and even that misses the mark. My question is for the experts; How would you approach these situations where your design requires priority for the weakest link in a high elevated, high Tx power deployment? 

PS. 2.4 is not an option due to BT headset interference.

[patoberli]

The typical suggestion, do a site survey has already been done. So the next best thing might be external antennas, but for whatever reason you already have decided upon internal antenna models?!? 

So in the end, discuss this again with the customer (?) and tell him about the downsides and limitations of the requirements. 

[littleyoda]

Well first reason is cost. The second and more important reason is the chip shortage and timeline for delivery. We don't have enough AXE to meet the requirement. 

[Leo Laohoo]

---

@littleyoda wrote:

Well first reason is cost.

---

Warehouse deployment is THE most difficult wireless configuration. 

It might be "extreme" but I always tell people "if cost is the reason why a proper wireless site survey cannot be done then DO NOT BOTHER PUTTING WIFI".  

Correcting a badly deployed WiFi in a warehouse costs about four times (conservative) as the original budget.  

[littleyoda]

Leo I completely understand. I'm going to try and revisit the issue with the branch to at least hold off for externals. I know however I will not be able to drop mount. For weakest link consideration, what antenna would you recommend at that height?

[Leo Laohoo]

---

@littleyoda wrote:

For weakest link consideration, what antenna would you recommend at that height?

---

Same as before:  Get a wireless site survey done.  

Only a site survey will be able to determine, with a very high degree of accuracy, what AP to use, installation height, what antenna, etc. 

Playing "20 questions" on paper is very different when the installation spectacularly fails (and someone is looking for someone to blame).

[patoberli]

Then you're better off delaying the installation until the parts are available. A shitty working solution can cause even more problems than the old solution that hopefully can work for a few more months.

[littleyoda]

I did the passive survey of what's existing installed. The coverage was moderate with Motorola APs at full transmit power which doesn't help my case other than it tells me I will need to design, whether it be a patch external or an omni internal, for full transmit power meet the -67 requirement for barcode scanners/PDAs. Remember my limitations. Installation has to be installed at the i-beam or at best on the wall. A high gain patch antenna will suffice but even so, with a -10dB offset, I still have to plan very densely to meet the requirement for lets just say a 14dBm deployment. So the conundrum I face designing at the moment is striking a balance between a very dense deployment / medium powered / increase probability of interference vs low dense deployment / max power / less probability of interference and no wiggle room.

An example for (3) 8 dbi antennas installed at ceiling 32 feet high at 14dBm with the -10dB offset added.

Surveyed (6dbi omnis @ max Tx)	Design (8 dbi patch @14dBm)

[littleyoda]

@Leo Laohoo do you have any advice for the above?

[Leo Laohoo]

---

@littleyoda wrote:

A high gain patch antenna will suffice but even so, with a -10dB offset, I still have to plan very densely to meet the requirement for lets just say a 14dBm deployment.

---

Will not work.  Does the wireless client have any "high gain antenna"?  They do not have.  So the AP is blasting the signal and, because the wireless client does not have a very strong antenna, the reply (from the wireless client to the AP) does not reach the AP. 

Another thing missing is the location of the aisle. How is the AP going to get positioned from the roof, etc. 

There are just so many factors to get a wireless AP deployed WRONG and this is just one of it.  

Finally, I am no longer answering any more questions (or sharing my opinions) about this thread.  

[littleyoda]

Please don't reply at all if you don't relatable input and shared experiences. I can't mark any of your replies as solution if they're not an answer to the problem. I was looking for professional replies with people who have experience in this category.

I'm aware of the client to AP balance. But there are environments that have design restrictions due to operation limitations. There's no perfect answer for this. I'm just looking for what other professionals have done in similar situations.

---

---

[patoberli]

In most cases they told the customer what must be changed to reach a well working solution. Of course, there rarely was such a shortage of devices like there is now, so this will delay many things. In any case, you already have the solution, but it doesn't make the customer happy. That's the sad thing about designing, you make a design that should work and provide it to the customer. If the customer then prefers "round windows inside square frames" then it's not the fault of the designers if it will not work. This is important and if you are also the one for the installation, to clearly mark this in the contract the customer must sign. In other words, make the customer sign a contract that tells him, that the design he prefers will probably not work and that you take no responsibility for it.

[Tony Rosolek]

Hello littleyoda,

I think that 32 feet is to high for regular/integrated omni directional antennas. So as you stated yourself some directional antennas like patch would be the right choice. 

But some questions: 

- how many clients do you expect on these 20-30 access points?

- where does the 10 dB offset come from? did you measure it by yourself? And where do you use it? in comparison to some specific measurement adapter or in your theoretical planing? (I'm asking because in theoretical Ekahau designs I would often not add this offset. I only use it with "real" practical measurement data. Otherwise you're planing for -57 dBm in 5 GHz which is quite aggressive.) 

- what barcode scanner model is used?

- why do you want to go down to 12.5 mW with the transmit power? Is there a specific device with requires such low power? I think 25 or 50 mW shall be fine as well. 

- is wall mounting an option?

[littleyoda]

Hi Tom,

Client usage is small. Probably no more then 5-10 devices maximum at a given point in time. The 10dB offset actually came from when I was using the Ekahau sidekick. I placed the TC51 on the iPad as I walked and surveyed, using WiFiMan android app. More often then not the signal was about a 8-12dB difference. I scorged FCC documents with Zebra on the device but no luck. I'm waiting on Zebra to get me the actual Tx power. I agree, not to trust the Ekahau built-in barcode offset. But it was pretty close. 

It's a Zebra TC51 which is 2x2 802.11ac. 

12.5mW definitely isn't doable and I agree that 25mW will be fine. I don't want to go too high because I need leverage to remediate any issues. From my FCC research, the RF scanners seem to range 13-15dB Tx power so I was trying to stay as close to that as possible. But yes 25-50mW would work.

Wall mounting patch antennas are an option and save a lot more in costs but you still have the weak-link problem. Overall, per the current design with 24 802.11ac Motorola APs at max power -- there's no issue. So I think my design with omni at the 30 feet mark, in double capacity is my best option. These situations are difficult and most often, sacrifices have to be made. So I had to choose between wall mounted patch or ceiling internal omnis. Because the feedback is there is no issue with the current design on these devices, it makes sense to just fill the gaps in.

I was just curious if anyone in the industry has this issue with these low powered devices. 

[Tony Rosolek]

Did I understand it correctly that there will be a maximum of 10 devices in the entire warehouse? Spread over more than 20 APs?

Are barcode scanners the only devices which will be used in the building?

I also did not found the transmit power for the TC51 scanners, but I think it shall be at least 50mW. 

What do you mean by "weak-link problem" related to wall mounting? 

Your offset measurement sounds reasonable. And it makes sense to use this offset when you compare it to Sidekick measurements. But I would not use it during theoretical planning with Ekahau.