315/433mhz signal strength indicator / discriminator

Some projects annoy me just because they should be easy and I can’t manage to complete them for some unknown reason. This one is not the longest time between shelving a project and completing it, but it certainly is the least work I’ve had to do to get it completed.

CPQ14USB uses the usb port shells in the current path for usb power

Side story, because apparently I didn’t blog about this when it happened. Back in high school I was working with a friend to cram a usb hub into a mouse. The idea was that we could keep a flash drive full of games or other data hidden in the mouse and it would be present as soon as we plugged in the mouse. After stripping the ports off the usb hub and adding wires to solder the usb flash drive directly to the hub board the whole thing wouldn’t power up. What I discovered maybe ten years later was that the super cheap single sided pcb for the usb hub used the mechanical shell of the usb ports as jumpers to complete the circuit for the negative wire from the incoming connection to the computer. That means that by removing the ports I broke the ground connection to the computer. That holds the record for being the longest it took me to solve some simple issue, but I’m glad I did and now I will never take for granted how someone else should have laid out their board.

unreasonably expensive for what it is

Ok, back to the main story. When I worked at GM in keyfobs we had a couple RF ‘sniffers’ that were used to detect whether a key fob was 433Mhz or 315Mhz. We could crack them open and look at the part number to determine that but this was faster and also established basic functionality (battery not dead and processor doing something). The meters we used were these (RF Solutions 006 signal strength meter) and these (RF Solutions 006-433 singal frequency signal strength meter). Imported from the UK, only displays one frequency’s signal strength at a time, and cost 100 GBP each plus shipping. Obviously being a multi-billion dollar company they wouldn’t buy one for each person to use, we had to spend time tracking each other down and borrowing/returning them because everyone needed them, just hopefully not all at once. I figured that I could just make my own so as to not need to borrow one all the time (and of course showing off was always the goal).

finished product

What do I need to accomplish my task? It needs to be pocket sized. USB rechargeable would be a nice upgrade over the 9v batteries the older ones just chugged (no, of course there were no batteries available at work, you were supposed to spend $10 in labor writing an expense report to get reimbursed for floating the company a $3 battery). Being able to display both bands at once would also be good since the existing meters either needed to switch between bands or only did one. Something analog to be able to discriminate between noise sources that may be causing interference and actual signal. You’d be surprised how much crap is in those cars that generates enough noise to swamp out the key fobs.

The first thing I needed was a couple of receivers that had RSSI out and not just a digital signal. It didn’t have to be calibrated to anything, but being able to get analog signal strength was all I cared about. That came in the form of these two from sparkfun (the RWS-374-3 and RWS-371-6). They report as having a ‘linear out’ pin as well as a ‘digital out’ on the datasheet. There’s not enough smarts in them to produce voltages outside the power rails of the module so I knew the range wouldn’t be an issue for the next part.

Next I needed to be able to convert that analog voltage out into a nice bargraph display. Say hello to our old friend the LM3914, also courtesy of sparkfun. These chips are great because they have LED current drivers as outputs so you can mix and match your LEDs and you don’t even need any resistors for them. I configured mine for bargraph mode instead of dot mode and broke out the high and low reference es as well as the LED brightness as trimpots so I could just tune in the whole circuit myself. I added series resistors to the LED current setting pots to give a sensible lower limit of brightness.

The last part I needed was the rechargeable battery, that was just a standard little usb lithium charge controller and boost circuit combo board and a salvaged lithium cell that fit the case I found. The power switch came from the pile of parts (super nintendo power switch?) and the case came from the pile at i3detroit. The case already had a circular hole and I wasn’t feeling fancy enough to make another faceplate that covered that hole so it remains. Also, I’m not sure if it was extremely good planning on my part that made the PCBs fit perfectly in the case between the standoffs but it totally works perfectly. At least part of me wanted to finish this because the PCB and wiring look so pretty and were assembled so well. There wouldn’t be another use for any of these parts, too much effort to salvage anything, it’s this or the trash bin.

it works!

So, when I returned to this project all I had to do was follow the tutorial and connect pin 8 to ground (already done on one side) and connect the inputs to the linear out on the reciever modules. It just worked, I don’t know what was stopping me before. I just tweaked the low setting on the bargraph drivers so ambient noise didn’t light up any LEDs and the high setting I think is just maxed. The brightness is also at the minimum I set with my fixed resistors and everything looks great.

RF Explorer, handheld spectrum analyzer

So, that’s what I couldn’t get done while at GM. Here’s what happened after I shelved it. One of the guys in the service department that spent a lot of his time debugging RF issues had himself one of these (RF Explorer WSUB1G) and compared to what we had it just blew us out of the water diagnostically speaking. Eventually through borrowing it and months of badgering by me and anyone else that got a chance to use the one I borrowed our group purchased two of these. They still had to be lent out and one for each team member would have paid for themselves in a month with man hours lost to returning and borrowing and tracking them down but whatever, it’s something.

EL-52545 TPMS and RF tool

This next one happened only after I left key fobs and was transfered to tire pressure sensors. The EL-53545 is the culmination of over a year of work between myself, certain members of the body electrical engineering team, the service department, and ATEQ. I could spend hours talking about all the work that went into this tool, but suffice it to say when that article I linked says “Thanks to Bob Wittmann” I can attest that it wouldn’t have gotten done without his help. This tool replaced the aging J-46079 tool that used an atmel processor that was end of life and they could not build any more without re-engineering it anyway. I helped convince them to add a general purpose signal strength test to the tool which would mimic the functionality of the RF Systems tool up at the top as well as decoding the VIN stored in the key fob if it had been programmed to a vehicle. Both of these would give the dealerships much better diagnostic tools and allowed us to write service instructions where they could actually diagnose RF issues with the vehicle.

Labeled (based on the FCCID of my key fob)

Would these last two things have gotten done if I had just finished my little signal strength meter when I needed it? Probably, but with how much work it took to get those done I hate to think I might have given up if I didn’t need them so badly to get my own work done.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: