As for voltage drop how low does your battery voltage go before the unit stops transmitting?

At a ~15 ma draw when transmitting the 2200 uf capacitor should easily give you several seconds of transmit time before its charge voltage drops too low. If not then the resistor could be reduced to a 100 - 200 ohm unit and a larger capacitor could be put in.

I'm not sure what is the cut off voltage before the remote stops working. I might be able to test that out with a variable DC supply.
If I recall correctly, the typical (average) transmit current is 7mA and the peak is 15 mA. Assuming I use 7mA and a 500 ohm resistor, the voltage drop across the resistor is 3.5V. That would leave me with 9.1V across the capacitor if the source battery is at 12.6V. Is this correct?

By the way, what is the capacitor for?
What could be the side effects if I don't use a capacitor?

Thanks again.

 

If your current limiting the spike then your right. Probably OK.

Littlefuse says around -100 V for 200 mS at each turn off. Significant? IDK.

ESD events are infrequent and random an > 15 kV and < 50 nS.

Pick how important it is. The TVS can still be used with the Zener regulator too.

Is this negative spike talking about connecting it up incorrectly or solely for ESD events or both?
Does ESD always give only negative spikes?
Assuming I choose to add TVS, where should I connect it?
Thanks again.

 

By the way, what is the capacitor for?
What could be the side effects if I don't use a capacitor?

The capacitor is there to work as a small battery to keep the voltage up for a few seconds while it's transmitting It als0 works as surge suppressor of sort being it has to charge up in order for the voltage to rise and the larger the value of the capacitorr the slower its rise time will be.

Might be a device worth reading about.

 

The capacitor is there to work as a small battery to keep the voltage up for a few seconds while it's transmitting It als0 works as surge suppressor of sort being it has to charge up in order for the voltage to rise and the larger the value of the capacitorr the slower its rise time will be.

Might be a device worth reading about.

Noted. Will google it.
Instead of using the remote button to activate, I think what I am planning to do now is to bypass the remote button and put a momentary push switch in front of the resistor. I am not sure if this configuration will take too long for the cap to charge up instead of it being always connected to the battery. Potentially, will this setup cause any issues or there is no difference with this setup as compared to not bypassing the remote button?

 

One thing to think about is "rolling codes". See: https://en.wikipedia.org/wiki/Rolling_code

The negative spike of around -100 V is part of the alternator regulation. Usually the rotor is pulsed to achive regulation. Each time it turns off, you potentially generate a spike. They are FREQUENT.

ESD or ElectroStatic Discharge in a normal device would be the human body collects a charge (the wool rug thing) (Sliding across a seat of the appropriate material). You touch the button and that charge gets into the input circuit of the device.

Many devices have ESD protection diodes in them.

 

One thing to think about is "rolling codes". See: https://en.wikipedia.org/wiki/Rolling_code

Sorry I must have not explained clearly.
I don't mean to bypass the entire remote control but just to bypass the remote push button (short it) and put my own push button in front of the resistor. Will that affect the rolling codes? Does having a standalone push button in front of the resistor not work the same way as if I am pushing the original push button on the remote?

 

Instead of using the remote button to activate, I think what I am planning to do now is to bypass the remote button and put a momentary push switch in front of the resistor. I am not sure if this configuration will take too long for the cap to charge up instead of it being always connected to the battery. Potentially, will this setup cause any issues or there is no difference with this setup as compared to not bypassing the remote button?

I don't see any reason make it more complicated than necessary. Just wire it up and push the button as normal.

The negative spike of around -100 V is part of the alternator regulation. Usually the rotor is pulsed to achive regulation. Each time it turns off, you potentially generate a spike. They are FREQUENT.

I have serious doubts about that. Modern vehicle systems are pretty well protected now given the huge amount of electronics they have plus any decent alternator has more than enough built in spike suppression to handle anything it could create itself when working as designed.

Besides how is a -100 volt spike going to make it past the alternators built in flyback diode in its regulator that is specifically there to prevent the rotor from making high voltage spikes plus output diodes and the huge current source/sink battery?

 

Sorry I must have not explained clearly.
I don't mean to bypass the entire remote control but just to bypass the remote push button (short it) and put my own push button in front of the resistor. Will that affect the rolling codes? Does having a standalone push button in front of the resistor not work the same way as if I am pushing the original push button on the remote?

I don't know if you have rolling codes or not. I don't know how to tell. I THINK you have to sync the xmitter and receiver when you have rolling codes, but not sure. If you turn on/off the remote say by a switch before the battery clip (resistor) and have the button pushed it might work or it might not. Self-tests during power up may see the button pressed and not work (doubt it). There may be a longer delay. It might not work.

You can't assume the method will work. Test first.

If you place a switch in series with the resistor that charges the cap, it will take 5 time constants (5*R*C) for the cap to be 99% charged. C in Farads. 22 uf = 22E-6 Farads.

 

I don't know if you have rolling codes or not. I don't know how to tell. I THINK you have to sync the xmitter and receiver when you have rolling codes, but not sure. If you turn on/off the remote say by a switch before the battery clip (resistor) and have the button pushed it might work or it might not. Self-tests during power up may see the button pressed and not work (doubt it). There may be a longer delay. It might not work.

You can't assume the method will work. Test first.

If you place a switch in series with the resistor that charges the cap, it will take 5 time constants (5*R*C) for the cap to be 99% charged. C in Farads. 22 uf = 22E-6 Farads.

I have tested the external push button switch without the resistor, capacitor nor zener and it works. I shorted out the remote push button switch.
Yes, I am concerned that it will take 5 time constants (or too long to be practical) so I'm thinking of not putting the cap and just the zener and resistor. Or using a cap with very low uF.
I'm not sure if not having the cap will have serious side effects other than not having some safety cushion of a surge suppressor.
Due to my lack of knowledge, this is all very experimental to me with no real understanding.
I'm going to measure the transmit current next to see if I can work out the value of the resistor to use.
I think there is rolling codes based on - what I heard about NICE brand.
Thanks.

 

I have tested the external push button switch without the resistor, capacitor nor zener and it works. I shorted out the remote push button switch.
Yes, I am concerned that it will take 5 time constants (or too long to be practical) so I'm thinking of not putting the cap and just the zener and resistor. Or using a cap with very low uF.
I'm not sure if not having the cap will have serious side effects other than not having some safety cushion of a surge suppressor.

If it works and you are going that route you could drop the capacitor value down to 500 - 1000 uF and the resistor down to a few tens of ohms.

Here's a great website with calculators to help you out.

http://mustcalculate.com/

 

For your understanding, The resistor limits the amount of current - hence limits the amount of damage a spike can do. Just like they say "current kills". Many IC inputs have to be current limited if the device is not powered and some external voltage is connected to it.

The capacitor acts like a battery for the duration of the remote button press. It can deliver a higher current than the resistor limited one for a brief time.

 

If it works and you are going that route you could drop the capacitor value down to 500 - 1000 uF and the resistor down to a few tens of ohms.

Here's a great website with calculators to help you out.

http://mustcalculate.com/

Yes, it looks like I will be going the route with external push button though it's getting more complicated than it needs to .
I plugged in some numbers into the calculator and getting some 'strange' large numbers. I think there is something wrong with my logic so i still need to check it out. Thanks a lot .

 

For your understanding, The resistor limits the amount of current - hence limits the amount of damage a spike can do. Just like they say "current kills". Many IC inputs have to be current limited if the device is not powered and some external voltage is connected to it.

The capacitor acts like a battery for the duration of the remote button press. It can deliver a higher current than the resistor limited one for a brief time.

That's the dilemma I'm facing, assuming I'm talking sense below.
I carefully measured the transmit current again and this time I'm seeing a 17mA current.
So, assuming worst case the battery is at 12.6V and 12V goes to the zener, I am left with 0.6V across the resistor.
If I need 17mA (and assuming no current flows in zener), then the resistor value must not be higher than 35 ohms, which is a small value and I wonder if can do any kind of protection against spikes. Regardless, a 35ohm resistor combined with 500uF gives me a 5 times RC constant of 0.9 seconds which I think is acceptable? My priority has to be that the remote has to work so I cannot have too high a value for the resistor.
What also puzzles me is when I push the external button, as it charges the cap, at some point in time, there will be enough voltage to trigger the remote. So, even before reaching 12V, let's say 11.8V, the cap will discharge. (I have tested it will work at 11.8V with a variable DC supply)
I also try to figure out what is the discharged current to the remote and this is where I'm lost.
I googled and saw i = c dv/dt and assuming dv/dt is a straight slope, I still cannot figure out how to find dv/dt.
I looked at data sheets and couldn't find any curves.
Normally, the remote will draw 17 mA from a battery but with a capacitor supplying the current, will this also be 17mA, or higher or lower?
Is the capacitor discharge current dependent on the impedance of the remote?; i.e. the dv/dt slope will vary depending on the capacitance I choose but the net effect is that I expect 17mA to be the output regardless of the capacitance value I choose (within reason).
If it is not and is much higher than 17mA, it might damage some components on the remote?
I know I am making this more complicated than it needs to be

 

I'm going to take apart pieces of what you said:

That's the dilemma I'm facing, assuming I'm talking sense below.
I carefully measured the transmit current again and this time I'm seeing a 17mA current.

Carefully: Probaby not. Why? If you used the current scale of your meter, it likely has a voltage burden. They are all over the map and sometimes not specified and depends on the range selected.

So, assuming worst case the battery is at 12.6V and 12V goes to the zener, I am left with 0.6V across the resistor.

Worst case is the car batter is 13.8 V and 18 V under a fault condition.
In designing a Zener regulator, you basically design the input current to be higher than the needed current.
The drop across the Zenier is used to check the power dissipation.

If I need 17mA (and assuming no current flows in zener), then the resistor value must not be higher than 35 ohms, which is a small value and I wonder if can do any kind of protection against spikes.

No! Here http://www.mouser.com/pdfdocs/ELNACalcuDYNACAPDischargeTime_Jul2011.pdf are calculations which are shown for super caps. e.g. Farads or close.

That 17 mA can be taken from the CAP. V0 and V1 are the charged cap voltage and the minimum the remote control will operate at,. the load won;t be constant either and you have some charging because of the resistor.
So, basically the longer you press the button, the less likely it is to work. rapidly pressing the button before 5 time constants are up, the cap isn't fully charged.
[/quote]

Regardless, a 35ohm resistor combined with 500uF gives me a 5 times RC constant of 0.9 seconds which I think is acceptable? My priority has to be that the remote has to work so I cannot have too high a value for the resistor.

it's more dependent on:
case 1 (connected all the time)
Your hoping the the transients get damped by the resistor and Zenier.
You need 5x time constants to fully charge
Usefull time can be estimated t = (C * (12-V1)/17e-3
if V1 = 5 then t = (500e-6 * (12-5)/17e-3 then you might get about 500 mS of use. It will be greater because your also charging. The 17 mA is probably operating mostly the LED. You can make that efficiency by a factor of 2, by using a 2 mA LED. With some tricks, i.e. replace the LED with an opto-isolator and drive the LED from the 12 V side knocks down the power consumption quite a bit.

Case 2: (button pushed all of the time & cap and resistor always connected)
You can, still just charge the cap all of the time and then power it with a switch from the battery terminal to the cap. Then it pretty much reduces to case 1. Each time you power up the remote, your introducing a transient.

case 3: (button pushed + switch before cap/diode combo)
In this case the cap has to charge. The current in a capacitor cannot change instantaneously.
Rolling codes might get messed up. Not recommended.


So lets just say that you make 5 * R * C equal to 10 seconds. it just means that you can't use the gate for 10 sec after you turn on the car.

Do add the reverse biased diode.

 

If I need 17mA (and assuming no current flows in zener), then the resistor value must not be higher than 35 ohms, which is a small value and I wonder if can do any kind of protection against spikes. Regardless, a 35ohm resistor combined with 500uF gives me a 5 times RC constant of 0.9 seconds which I think is acceptable? My priority has to be that the remote has to work so I cannot have too high a value for the resistor.

What calculator are you using? On that site I linked to a 500 uF capacitor and a 35 ohm resistor give a~40 mS (~.040 second) time constant.

 

What calculator are you using? On that site I linked to a 500 uF capacitor and a 35 ohm resistor give a~40 mS (~.040 second) time constant.

my own calculator. I used example 3 template in the site under capacitor charge and discharge and it asked for From and To voltage which I don't know what to put. I tried from 12.6V to 4.15V (33%)? The results give a peak current of 360mA; way more than the 17mA needed and that is why I asked whether the discharge current is a function of the impedance of the remote. I thought my inputs cannot be right.

So I took 500u X 35 and then multiply by 5 to get "5 RCs" of 0.9 secs.
May I ask what are the parameters and numbers you used to get the 40mS?
Thanks.

 

it's more dependent on:
case 1 (connected all the time)
Your hoping the the transients get damped by the resistor and Zenier.
You need 5x time constants to fully charge
Usefull time can be estimated t = (C * (12-V1)/17e-3
if V1 = 5 then t = (500e-6 * (12-5)/17e-3 then you might get about 500 mS of use. .

I don't quite understand this but I think to simplify things, it makes sense to have the external push switch, S2 to be after the cap. That way, the cap will be charged all the time until I push S2. I missed out the 35 ohm resistor in the drawing.

So lets just say that you make 5 * R * C equal to 10 seconds. it just means that you can't use the gate for 10 sec after you turn on the car..

LOL. If this needs to end up waiting 10 secs then its game over. At most, I think 2 secs wait time is max.

Do add the reverse biased diode.

I'm not sure how to connect this. I saw on the data sheet that it says normally its connected in parallel with the device i.e. D2 in the picture. I cannot figure out how that works against a negative 100V transient I shown in red. So I show another position to connect as in D1. Which is correct or are they all wrong?

One more thing regarding this negative voltage to help me understand further if you don't mind.
Suppose I am able to borrow an AC powered oscilloscope 50MHz, 500 MSa/s.
I set the probe at 10X and connect the positive to the battery terminal and negative to car chasis.
Assuming I know how to set trigger levels, time base and Volts/div etc.
I then start the car for say 20 secs and then turn off the engine and repeat this 10 times everyday until a transient happens.
My question is not whether I can see that negative or positive transient.
My question is whether what I described above is one valid method to 'see' or measure that transient?
Thanks

 

my own calculator. I used example 3 template in the site under capacitor charge and discharge and it asked for From and To voltage which I don't know what to put. I tried from 12.6V to 4.15V (33%)? The results give a peak current of 360mA; way more than the 17mA needed and that is why I asked whether the discharge current is a function of the impedance of the remote. I thought my inputs cannot be right.

So I took 500u X 35 and then multiply by 5 to get "5 RCs" of 0.9 secs.
May I ask what are the parameters and numbers you used to get the 40mS?
Thanks.

That online calculator link.
It even gives you pretty little graphs to show you what's going on during the charge cycle.

 

D2 is in the wrong direction.

D1 is OK to use, but not strictly required. It does offer easy reverse polarity protection.

 

Z12 won't like having the full amp capacity of the vehicles system applied to it either.

I also don't see the point of shorting out the original switch and adding an extra one to it being the stock switch is doing what the add on external one does anyway.

Personally, I think this is getting way over engineered for what should be a simple concept.

I'd still like to know where a -100 volt surge is going to come from that can get past the alternator's diodes plus the 1000+ amp dead short capability of the battery plus all the other surge suppression built into the vehicles electrical system that could shoot through a 50 -100 ohm resistor drain a large capacitor and overdrive a 12 volt Zener diode in its forward conduction mode to the point of being able to damage the transmitters circuitry without wiping out the vehicles whole electrical system?

I doubt I could do that using a junkyard cranes 2 ton 30 KW electromagnet if I tried channeling its inductive kickback energy through a vehicles electrical system on purpose.