Key fob only works on battery and not power supply. What?

[thatsmessedup]

TL;DR Summary

Key fob only works on battery and not power supply. What's going on here?

Yesterday the batteries died on my key fob and I didn't have the appropriate batteries to replace them. I thought I could use my power supply to operate the key fob to open the car door but it did not work. I thought that it may be caused by a noisy power supply at the higher frequency range of 432MHz but when I got my hackRF to investigate the waveform with the battery and then with the power supply, I was shocked to find no signal at all. I have since replaced the batter and everything works great, but as a recently graduated EE I am at odds with what is going on. Can anyone help me figure out this mystery. Here is a youtube link of the issue

 

[berkeman]

Any chance you connected the PS with reversed polarity? (Although I'd be surprised that it didn't blow up the fob) Full disclosure -- I didn't watch your video.

Also, what did your PS show as output current?

 

[thatsmessedup]

Polarity is good and I tested to make sure it was 6[V] with a multimeter. current is too small to measure on the power supply so I don't know off hand. I suppose I could measure it if it will help.

 

[berkeman]

What current did your power supply show?

 

[thatsmessedup]

What current did your power supply show?

current is too small to measure on the power supply so I don't know off hand. I suppose I could measure it if it will help.

 

[berkeman]

Trasmit current will be measurable, no?

 

[thatsmessedup]

Trasmit current will be measurable, no?

I have measured the current to be 5.5-6.5 [mA] when the button is pressed

 

[hutchphd]

Is the current limit on the power supply set high enough (do you maintain full 6V at transmit?). Are the multiple pads (square array) on the negative battery pad there for a functional reason? Must admit I'm flummoxed...grasping at straws

 

[thatsmessedup]

I cranked up the current limiter and also tried a different variable power supply with no luck. I can't imagine that the grid is anything more then to make sure the battery is touching.

 

[thatsmessedup]

This is the power supply I'm using.

 

[thatsmessedup]

 

[thatsmessedup]

Here are some closeups of the two major chips on the PCB.

 

[hutchphd]

Is the battery used as the antenna somehow? (I see no real antenna.)

 

[Rive]

The thing has no sufficient buffer caps. Through a battery this is not a problem, it can support some silly big current peaks on short term: but a PSU wiring will prevent going through any HF current, so no power for transmit.
Add some 100uF tantal (or: ceramic - a miracle that you can have these in ceramic!) to the battery terminals when it's with a PSU. I think it'll work with that.

 

[256bits]

Is the battery used as the antenna somehow? (I see no real antenna.)

the looped trace in post 9 , i presume.

 

[thatsmessedup]

The thing has no sufficient buffer caps. Through a battery this is not a problem, it can support some silly big current peaks on short term: but a PSU wiring will prevent going through any HF current, so no power for transmit.
Add some 100uF tantal (or: ceramic - a miracle that you can have these in ceramic!) to the battery terminals when it's with a PSU. I think it'll work with that.

That makes a lot of sense. Maybe I can give this a try at some point but right now I don't have one of those on hand. Thanks!

 

[Tom.G]

Your video shows it working with the battery and you holding it with one hand.
With the power supply, you are holding it with two hands, left hand to push the button and the right hand holding the clip lead on the battery contact. The index finger of your right hand is contacting the antenna loop around the edge of the pc board, thereby greatly loading and detuning the antenna; probably enough to to disable the xmit oscillator.

A method for power connection is to insert a piece of paper and a piece of Aluminium foil between the circuit board and the battery, foil against the board, paper to insulate the battery. You can then clip the power supply lead to the foil without frying the battery, while still powering the board.

Cheers,
Tom

 

[thatsmessedup]

I think adding the capacitor will make it work. One last thing I tried was smaller leads but I think they were still to long. The wavelength at 432[MHz] is about 70[cm] so the leads would need to be much smaller than that to have a chance at sourcing power in a timely manner to be transmitted. Additionally, even a few [nH] of inductance will create a lot of reactance at these high frequencies. So, I'm hoping that a low inductance, close proximity source will operate the key fob. I've got some tantalum caps on the way... We will see. Thanks for all the helpful tips and lessons and ill keep you updated if the caps do the magic.

 

[pbuk]

The last picture in this post shows a CR2016 which is a 3V battery.

[Edit:]But the video shows that there two of them so that is good (CR2032 equivalent).

 

[hutchphd]

[Edit:]But the video shows that there two of them so that is good (CR2032 equivalent).

(forestalling unfortunate incidents)

The CR2032 is a 3V lithium single cell just twice as fat . Two CR2016 in series are not equivalent so I am not sure what you are saying

 

[pbuk]

(forestalling unfortunate incidents)

The CR2032 is a 3V lithium single cell just twice as fat . Two CR2016 in series are not equivalent so I am not sure what you are saying

Nor am I - edited.

 

[NTL2009]

I think adding the capacitor will make it work. One last thing I tried was smaller leads but I think they were still to long. The wavelength at 432[MHz] is about 70[cm] so the leads would need to be much smaller than that to have a chance at sourcing power in a timely manner to be transmitted. Additionally, even a few [nH] of inductance will create a lot of reactance at these high frequencies. So, I'm hoping that a low inductance, close proximity source will operate the key fob. I've got some tantalum caps on the way... We will see. Thanks for all the helpful tips and lessons and ill keep you updated if the caps do the magic.

Have you tried the suggestion in post 17? You can test if hand position is affecting the antenna while you are waiting for the caps.

 

[thatsmessedup]

The thing has no sufficient buffer caps. Through a battery this is not a problem, it can support some silly big current peaks on short term: but a PSU wiring will prevent going through any HF current, so no power for transmit.
Add some 100uF tantal (or: ceramic - a miracle that you can have these in ceramic!) to the battery terminals when it's with a PSU. I think it'll work with that.

Gave it a try with a 100[uF] tantalum capacitor on the terminals. Not a single blip from the HackRF.

 

[Averagesupernova]

This is a pretty simple looking device as far as individual components go. Have you made a schematic and taken some voltmeter readings? Are you sure you are actually getting the unit powered?

 

[thatsmessedup]

This is a pretty simple looking device as far as individual components go. Have you made a schematic and taken some voltmeter readings? Are you sure you are actually getting the unit powered?

No I have not made a schematic, but I really thought the cap would make the fob do the magic. There are still many things that can be going on here. Possibly taking out the battery removed quite a bit of metal from inside the loop that is supporting the antennas resonance. At this point it would be hard to test as I need this keyfob to use my car and I'm a little scared ill do something stupid and damage the unit.

 

[thatsmessedup]

Have you tried the suggestion in post 17? You can test if hand position is affecting the antenna while you are waiting for the caps.

I tried different hand positions and there was no difference.

 

[DaveC426913]

I would like a pic of the OP lugging that bigass power supply around on his keychain

 

[tech99]

I would like a pic of the OP lugging that bigass power supply around on his keychain

It is possible that the power leads introduce too much radiation damping into the circuit. If it is a super regen receiver they are very sensitive to having too much loading.

 

[hutchphd]

With the fob on battery power and producing the pulse train can you threaten it with the power supply leads (one side at a time) and make it stop ??

 

[rsk]

Right, might be nothing to do with this, but as a high school physics teacher, we sometimes make electromagnets with a nail and a coil of wire.
As a young teacher, short of batteries, I assumed I might be able to make these using school power packs which look a bit like that supply you have there. But no, they made rubbish electromagnets, batteries are miles better.
At some point, maybe when teaching ac/dc to students, I realized that the trace on the dc supply is nothing like the battery, it's 1/2 wave rectified and not particularly pretty (yes, I know, this shouldn't have been a surprise to me).
I assume that was what made the made it fail as an electromagnet.
Is that why it's not working for you?

 

[hutchphd]

I assume that was what made the made it fail as an electromagnet.

I think it more likely for the electromagnets that the batteries were just more capable to produce high currents for short time intervals. Batteries are pretty impressive and an electromagnet is a current-hungry device. The half-wave sag will become more pronounced at high current too.

 

[rsk]

True, school power packs often state 1A max and flip off pretty quickly.

 

[CWatters]

Gave it a try with a 100[uF] tantalum capacitor on the terminals. Not a single blip from the HackRF.
View attachment 288192

When you remove the battery does the +ve battery terminal short the power supply by contacting the -ve terminal on the pcb? Especially if you have it the other way up to press the buttons. Try replacing the battery with cardboard when on the power supply.

 

[hutchphd]

When you remove the battery does the +ve battery terminal short the power supply by contacting the -ve terminal on the pcb? Especially if you have it the other way up to press the buttons. Try replacing the battery with cardboard when on the power supply.

The LED lights when the button is pushed.