Hey all. I’m redesigning my car keypad board (car was totalled, new car works differently) and I was wondering about the practice of using jumpers (like a hard drive or cd-rom has, not jumper wires). I want to make this board more universally usable for different cars, so I need a way to bridge/reroute certain traces on the board after the fact. This board is going into the door of a car though, so I wondered if anyone here could speak for their longevity with vibration and possible moisture. If they are a no-go, what alternatives would you suggest?
In my experience, jumpers are pretty good. Of course, I haven't tested them in the conditions you're talking about, but they are meant to hold up to a little bit more mechanical wear than bare traces. You could also opt to use switches, which can be very durable. Either way, consider installing your circuit into a weather-proof box of some sort, that will help quite a bit.
Jumper clips should work fine. If you are very concerned you could always dab a little hot glue (or RTV glue) to hold the jumper to the PCB, which is easily removable later if required.
End-user configurable, solder-bridges really hurt that.
My first board was custom made for my car only, all smd, no switches. I'm trying to cut a compromise between being universal, not requiring end user soldering (except adding resistors when needed, no way around that, I imagine different cars need all kinds of different values) and longevity (no switches failing, jumpers falling out from door slams).
I feel like switches will fail, but jumpers can vibrate off. What can be said for dip switch longevity? At least I wouldn't have to worry about them falling off.
I would rather require the end-use of a soldering iron before I use a component that will fail. I don't know how wet the inside of a car door may get. Convenience is nice, but reliability is paramount.
I'm flip-flopping back and forth on this alot.
Solder-bridges require less space and less parts($), also most reliable. Jumpers not very much more space, possibly vibrate off? Switches more space, more money, but also more convenient for end users.
--- Board explanation below, optional, for those who want some context ---
Basically my board has two solid state relays, one each for lock/unlock. My last car required me to close the same two wires together for lock/unlock, but through different resistances (330 ohm and 1K ohm). My new car requires me to close one common wire to two different ones for unlock/lock.
I'm sure that there are other types that require something else.
So I want a jumper/switch/whatever to optionally link the inputs if they are shared (so I don't need a jumper wire on the outside), same with the outputs (independent of input setting). On the output trace, there should be a spot to solder in an appropriate resistor. I may want a jumper/switch to bypass this with a solid trace, or I could just solder in a jumper wire in place of the resistor.
I don't think the 'end user' (the car owner) will actually be setting this device up, it seems more likely that the 'installer' will be doing this. As part of the installation you could have the installer temporarily connect a specific programming jumper and apply power. The software would detect this jumper and make some sort of change to a value stored in the internal EEPROM. A different programming jumper would result in different data being written to the EEPROM. When the device is later powered up without any programming jumpers the software would go to the EEPROM to determine the appropriate configuration.
Ah, I see. Your method would not rely on a permanent jumper, therefore nothing to wiggle off. Unfortunately, to have a software-selectable path after the
SSD SSR, I can only imagine accomplishing this by having more SSDs SSRs on the board. This will take up more space and cost more cash. I believe they are the second most expensive thing on the board (after the power module). I went with SSD SSR because I won't know exactly what I will be switching, depending on the car.
You're probably right about 'end-users' not actually setting this up. I forget that most people who have things added on to their cars aftermarket (keyless entry, etc) don't actually know anything about doing it, and just pay a professional. I'll be going with the solder-bridges then. I'm not planning on turning this into a business venture, I'll just be doing this for myself and my dad right now (he already has one on his Lincoln, wants it on his Dodge). But if a friend wants one, I want to be able to provide one without redesigning the board again. Since I will be involved with each install, soldering will be fine. The end result will not leave me worrying about it like the others would. Thanks for all the input guys.
Oops, brain-malfunction. I meant SSR, solid state relays, as mentioned in my prior post above.
neonpolaris: Oops, brain-malfunction. I meant SSR, solid state relays, as mentioned in my prior post above.
Most SSRs are for controlling AC voltage only, not a common application in an automobile?
I wanted to be able to switch a potentially unknown signal, most likely dc, but be indifferent of the direction of current. Would an SSR be bad practice in this situation? If so, what would be a better choice?
Edit: I think it was an AQW212A that I used on my first board. Link here: http://www.mouser.com/ProductDetail/Panasonic-Electric-Works/AQW212A/?qs=sGAEpiMZZMtLEhJ5P%2fNsZx%2fVqDTYcQwamYMh1RNGSig%3d
Would an SSR be bad practice in this situation?
The common SSRs (based on triads or SCRs) would not work for DC switching as once the device turns on it could not be turned off. The device you linked to will work for DC or AC as it’s a FET switch, however it’s 500ma max load current would limit it to what kind of loads you could control.
KE7GKP: Exactly what is your application here?
I'm simulating a door-lock button press on the driver's door of a car. I'm not directly driving the door locks, just signalling to the car in the same way as the button on the door does. On my first car, the physical switches in the door closed the same two pins on the wiring connector together with different resistors. My board tapped into these wires in the door, using the same value resistors located on my board. I suppose I could look at the max ratings on the lock button, but I doubt they are driving much. My first board used the above mentioned MOSFET-based SSR flawlessly for about a year before the car got totalled. (Link: http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1236735887/15)
I'm not saying that they are the best thing to use, but they seemed like the best to me considering my limited electrical knowledge. If anyone has a better suggestion, I'll gladly listen and read up on whatever topic I need to to better educate myself/make the board better. The thing to take away the most is that I don't know how every single car's door controls work, so I want something that will switch any type of signal no matter the direction of current. Size is important, and no moving parts would be preferrable. Again, I don't think that any modern (OEM!) door lock controls directly drive any door lock actuators, no I don't expect a lot of current, but I admit that this is an assumption.
Changes I'll be making from my original board linked above:
- ATmega328, power regulation, and supporting components will be integrated into the board, no third-party boards used.
- LED backlight will be driven from transistor instead of expensive SSR.
- Board will be designed to be more universal, though will not be designed to accommodate security systems of any type.