Maker Pro
Maker Pro

Help with time delay relay circuit

steamngn

Aug 27, 2012
22
Joined
Aug 27, 2012
Messages
22
Hello Electronics gurus!
I have a diesel tractor that has a magnetic reed switch in the fuel tank to indicate low fuel. This switch is open when there is fuel in the tank, and closes when the level drops below it. When this switch closes a 12v relay is energized to disengage the fuel control and shuts off the engine. The problem is, sloshing fuel causes the float switch to intermittently open/close, which shuts off the engine! What I am looking for is a time-delay circuit, so that when the float switch closes a delay of 10-20 seconds starts. If the switch STAYS closed longer than 20 seconds, the delay timer energizes the 12v relay and kills the engine. If the float switch OPENS during that 10-20 seconds, the timer needs to reset. My electronics experience is limited, and I have looked at perhaps a 555 chip to do this, but I cannot figure out the reset portion at all. Could someone show me a circuit diagram to do what I need? Any and all help is greatly appreciated!
Andy
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
Perhaps something like this would work.
Note that there are two extra gates in the 4093 which will need their inputs tied to something.
I am too tired to check the logic tonight.
 

Attachments

  • fuel 001.png
    fuel 001.png
    3.1 KB · Views: 896

steamngn

Aug 27, 2012
22
Joined
Aug 27, 2012
Messages
22
Hi Duke,
Thank you for helping.. hope you got some rest!
I will need some more detailed help though, as my electronics knowledge and ability aren't that good :eek: ! While I can understand a schematic and get something assembled, I really don't know that much about individual transistors and their abilities/functions... and so will need a pretty complete schematic to pull this off. Guess I need some skoolin'! :D
!!!
After some more digging I found the attached schematic;
does this fit my criteria correctly? At first glance it appears to... I added the transistor to carry the load of the relay, which I believe is .16 amps. I THINK I understand this! One question I have is:
does this keep the relay energized after the time delay, or is it just a pulse that will energize then de-energize? I would prefer it to keep the relay energized if possible...
Andy
 

Attachments

  • timedelaycircuit.gif
    timedelaycircuit.gif
    19.5 KB · Views: 4,075
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
In this circuit, the input will normally be low, Q1 will be turned on and the capacitor will be charged. Th 555 output will be low and the 2N3055 will be off.

If the input goes high for a short time, then the output will not change until the capacitor is discharged. Intermittent input will keep the 2N3055 off.
 

steamngn

Aug 27, 2012
22
Joined
Aug 27, 2012
Messages
22
Hey Duke,
If I am understanding you correctly, then this is just what I need. The input (float switch) bounces up and down as the tractor is moving, so the input will flip back and forth from low(open) to high (closed), and this needs to NOT set the output high unless the float STAYS high (closed, due to the fuel level being low enough).
When the input flips to low before the capacitor discharges, the output remains low and the capacitor will recharge, effectively resetting the time-delay correct?
So then this IS correct? :eek: I'm doing a little victory dance!
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Why is a 2N3055 being used to drive a relay coil? What's the resistance or current of the coil? I think I read .16A (160mA). Seems like a whole lot of over kill. For that matter, you don't need a transistor to drive a 160mA relay coil. The NE555 will drive it directly. If you're using it for the reason of low side switching I can appreciate that but not the 2N3055.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
After some more digging I found the attached schematic;
Andy

Where did you find that? It has multiple befuddlements.

By the way, I thought I made that word up but ISpell says it exists. Go figure!
 
Last edited:

john monks

Mar 9, 2012
685
Joined
Mar 9, 2012
Messages
685
Steaming, you might want to consider what happens when your diesel engine invests a large gulp of air. You might have a very hard time restarting it.

What is the equipment you are running? Maybe I can check.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The CMOS circuit which I showed earlier works in a similar way to the one you show. A diode is used instead of a transistor but otherwise almost identicle.
The CMOS circuit will have a couple of spare gates which could be used to trigger a warning when sloshing commences.

You will need to give details of the input and the output load if a final design is to be suggested. A 2N3055 may need a heat sink (depending on the load) and a FET may be a better option.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
You can do what you want with duke37's 4093-based design, or a design based on a 555 (not the one you've dug out - something much simpler will do), or a circuit using a couple of transistors. You would have to build it on a piece of stripboard. If you're powering it from an automotive supply, the circuit will need protection against surges on the power supply and noise on the input.

Do you have experience with electronic construction? Or a desire to get involved with it? Do you have tools? Or would a ready-made product be better?

Here's an eBay search results page:
http://www.ebay.com/sch/i.html?_nkw=12v+time+delay+relay

I THINK that the relay they call "H3Y" will do what you want. That number seems to be available from more than one source. I THINK the way it works is that when you apply power to it, it delays by an adjustable length of time (up to 30 seconds) before its contacts close, and if power is removed before the delay expires, the delay restarts from zero the next time power is applied. IF that is how it works, you should be able to use it. Once the delay has expired and the contacts have closed, as long as the fuel tank microswitch stays closed, the relay contact should stay closed as well.

I have not been able to find a description of how the relay works, apart from the fact that it's an "adjustable time delay relay". So I could be wrong. They're only around USD 5 so it might be worth taking a risk and buying one.

Here's another hit on an H3Y that includes a mounting base:
http://www.aliexpress.com/product-f...0-30-SecH3Y-2-Baseond-DC-12V-wholesalers.html

If you'd rather build something, tell us how it's going to be powered, so we can add protection to the existing designs. I can also suggest a 555-based design and a transistor design with circuit descriptions so you will have a basic understanding of how the circuit works.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Good Stuff!

Perhaps something like this would work.
Note that there are two extra gates in the 4093 which will need their inputs tied to something.
I am too tired to check the logic tonight.

Duke, this is wonderfully simple and effective. The logic is fine too! I spiced it using a timed float switch. Just needs some filtering and spike protection on the Vdd to handle the nasty environment. ;)

Chris
 

steamngn

Aug 27, 2012
22
Joined
Aug 27, 2012
Messages
22
Ok Gang,
Let me see if I can answer everyone's questions:
First, the circuit and requirements:
The unit is to be installed in a 12v automotive environment, so proper protection is required.
The fuel tank float switches are very low current, and it appears to me that an electronic circuit would be ideal (I could be wrong!). At any rate, here is the list:
  • 12v automotive power supply
  • If the float switch closes, a time-delay of 10-20 seconds needs to start.
  • If the float switch opens before the time-delay expires, then the circuit needs to reset
  • If the float switch stays closed, then a relay needs to be enegized
  • The relay coil load is ~ .16 amps @ 12v
I've looked at time-delay relays, but the floats are rated way lower than the coil loads, so I've decided an electronic circuit would be better.
As for the 2N3055, I added that after doing some reading and thinking that the 555 would last longer this way, but it is becoming apparent that I'm over-engineering this thing!
f you'd rather build something, tell us how it's going to be powered, so we can add protection to the existing designs. I can also suggest a 555-based design and a transistor design with circuit descriptions so you will have a basic understanding of how the circuit works.
And I actually HAVE a LM555 and an LM556 chip in my parts box, so lets go with that!
I have limited electronics construction experience, but I am capable and have the tools to assemble a simple circuit on a breadboard such as this, and would like to try. If you guys can provide me with the circuit I'm sure I can build it, and I will learn something along the way!
Thanks so much to everyone for their help!
Andy
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Thanks for the extra information.

What's the current rating of the float switch? Did it previously drive the relay coil directly?
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
And I actually HAVE a LM555 and an LM556 chip in my parts box, so lets go with that!
Andy

Having one isn't good criteria for choosing a component. Especially when a 4093 cost about $1.50 and does exactly what you want it to do. All it needs is a little 2N2222 or similar driver transistor.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Availability is always a consideration, but in this case he will have to order other parts as well.

In an automotive environment I would avoid CMOS if possible. Actually I'd avoid the 555 as well. Both of them have an absolute maximum supply voltage that's uncomfortably close to the working voltage, and they're easily damaged by overvoltage, so I'm including some fairly heavy surge suppression in the design I'm working on. Personally I would use a discrete transistor design.

The 555 can be used as a Schmitt trigger, and compared to the 4093 it's smaller, more robust, and doesn't need an external transistor - all significant advantages here. The 4093 is a great chip and has lots of applications, but I don't think it's the best fit for this one.

Andy, are you SURE the float switch can't switch enough current to drive the relay coil? Arranging the circuit as a time delay relay, powered from +12V via the float switch, would make things a bit simpler and tidier.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Here's my suggestion for a 555-based design.

attachment.php


Power from the automotive 13.8V rail and ground is applied between CN3 and CN6. R4 is a heavy duty spike suppression varistor designed for 14V operation, but its clamping voltage isn't low enough to protect the 555 so I've added a 3.3 ohm series resistor (R3) and a 15V 5W zener diode (D3), which also protects against reverse polarity. C2 smoothes the power supply rail.

At the other end, while the float switch is open, R1 pulls the CN2 voltage to 0V, which causes diode D1 to conduct and pull the C1 voltage down to nearly 0V. When the float switch is closed, it pulls the CN2 voltage up to the positive rail, and D1 no longer has any effect, so C1 is able to charge slowly through R2.

So while the float switch remains closed, C1 charges up, but as soon as the float switch opens, C1 is discharged rapidly through D1 and R1. C1 will only continue to charge if the float switch remains closed.

The 555 is used as a Schmitt trigger with a positive threshold of 2/3 of the supply voltage. When its input (pins 2 and 6, which are tied together) exceeds this voltage, it responds by driving its output (pin 3) low. This energises the coil of the relay. D2 protects the 555 against the back EMF spike generated when the relay coil is de-energised.

The relay coil must not draw more than 200 mA. That is the maximum rated output current of a standard 555. Also, don't use a low-power 555-type device (TLC555, TS555, ICL7555) as their output drive capabilities are lower.

The time delay is approximately equal to the product of R2 (in ohms) and C1 (in farads). For the values given, this is 680e3 * 22e-6 which is about 15 seconds.

All of these components are available from http://www.digikey.com.
 

Attachments

  • Steamngn.gif
    Steamngn.gif
    7.7 KB · Views: 1,355

steamngn

Aug 27, 2012
22
Joined
Aug 27, 2012
Messages
22
Hey KrisBlueNZ,
Perhaps it would help if I lay out the details of the fuel tank and entirely:
The tractor is a vintage 1969 machine that was originally gasoline engine powered.
The engine is a still-in-production Yanmar direct injection diesel.
I built this baby; attached is a picture of my mechanical work for your enjoyment :D!
The tank actually has two of these float switches in it vertically separated by 4 inches, and they drive a two-color led on the dash (Holy Crap! I never mentioned the led!!! :eek:). I do not know the exact current rating, but the wire is on 26 gauge so I'm sure will not directly power an automotive relay, at least not for long.
The lower float (the one we are working with for the shutoff) had a small transistor circuit so that when the float is up (open) the green led is lit. When the upper float is also up (open), the red led is off, and the fuel indicator is green. When the upper float closes and the lower float is still up the led is yellow (red and green lit). When the lower float closes, the transistor circuit switches off green led, and now the fuel indicator is red. This system was never designed to switch a relay, and this transistor circuit has recently failed (After 15 years!). Hence my quest for a replacement.

NOW,
The big issue here is that I need a way to drop power to the fuel shutdown holding coil; this is a higher-current load on the order of 9 amps @ 12v dc, and this load is constant while the tractor is running. I also need to compensate for the lack of baffles in the fuel tank which cause the floats to bounce up and down which will inadvertently kill the engine.
SO,
A time delay that resets and can power a small relay is perfect. The schematic I provided was only cobbled together from my limited knowledge of electronics; I liked the fact that it had an adjustable time delay, but that isn't a requirement.

Kris, Can the green led be added into your circuit so that when the float is open the led is lit? It can also be driven from the N/C circuit on the relay if not...
And thanks again for all the help! Great details and great conversation from everyone!
Andy
 

Attachments

  • tractor 001.jpg
    tractor 001.jpg
    162 KB · Views: 346

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
OK thanks for the description.

I'm no tractor expert, but that looks pretty tidy :)

I have a few more questions.

Are the float switches fully isolated from each other and from everything else on the tractor? So you just have two 2-core wires, one from each float?

The fuel shutdown holding coil needs to be normally energised, so when it's de-energised, the fuel supply is cut off, is that right?

Are you sure that it draws 9A at 12V? That's an awful lot of power. I doubt it will be that much. But even if your fuel cutoff relay draws only an amp or two, you should probably still have an interposing relay between the circuitry and that relay, for robustness.

Do you know the characteristics of the dashboard LED? I assume it's a 3-wire type. Is it common anode or common cathode? Do you know what current it is driven at? Do you have a schematic or photograph of the original driver board?

Do you want to keep the LED and keep the behaviour that you described?

Do you want to make a new board that drives the LED and provides the relay output to the fuel relay?

Yes, you can drive the LED from the float switch, with the addition of a buffer transistor. Would you rather have a beeper to warn you of low and very low fuel levels?

Just wondering... why do you need the fuel cutoff to be automatic and based on the float switch? Wouldn't an audible alarm be enough, along with a manual fuel cutoff?

How would you feel about a circuit with a few transistors instead of a 555?
 

steamngn

Aug 27, 2012
22
Joined
Aug 27, 2012
Messages
22
He ya go!
Are the float switches fully isolated from each other and from everything else on the tractor? So you just have two 2-core wires, one from each float?
Correct - both floats are fully isolated from everything and are two-core wires.
The fuel shutdown holding coil needs to be normally energised, so when it's de-energised, the fuel supply is cut off, is that right?
Correct. The fuel shutdown solenoid is actually two coils, one pull and one holding. De-energizing the holding coil kills the engine.
Are you sure that it draws 9A at 12V? That's an awful lot of power. I doubt it will be that much. But even if your fuel cutoff relay draws only an amp or two, you should probably still have an interposing relay between the circuitry and that relay, for robustness.
8.37! I know it is a lot, and it is possible that it is higher than it should be due to age or something. But since its that high and working, we should engineer accordingly...
Do you know the characteristics of the dashboard LED? I assume it's a 3-wire type. Is it common anode or common cathode? Do you know what current it is driven at? Do you have a schematic or photograph of the original driver board?
I do know the led specs, as I had to replace it once. It is 3 wire, 12v 20ma 2.2fv. I believe it is common cathode, but I don't remember! There is a 560 ohm resistor on each leg.
No pic or schematic of the original circuit, long since relegated to the trash bin...
Do you want to keep the LED and keep the behaviour that you described?
Do you want to make a new board that drives the LED and provides the relay output to the fuel relay?
Yes, and yes. Since the old board is dead and there was no shutoff control now is the time to improve!
Yes, you can drive the LED from the float switch, with the addition of a buffer transistor. Would you rather have a beeper to warn you of low and very low fuel levels?
Just wondering... why do you need the fuel cutoff to be automatic and based on the float switch? Wouldn't an audible alarm be enough, along with a manual fuel cutoff?
The led is highly desirable; A beeper that would sound BEFORE the fuel is cut off would be great! As for the auto-cutoff, running a diesel out of fuel is a bad thing! bleeding air out of an injection system even this small can take hours!
And while I try to monitor the fuel level, there are times where other people use the machine and they just aren't thinking about it, so I need to protect against this.
How would you feel about a circuit with a few transistors instead of a 555?
I'm open to just about anything, the 555 was just where I started out. The minimal complexity of the circuits so far is ok with me, but if you have a better idea then educate me! I'm ready to try and build whatever comes about, it is a way for me to learn!
Andy
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
OK, here's an updated schematic.

attachment.php


This design uses two transistors instead of the 555 and it doesn't need as much protection against spikes on the supply rail. I've added the second float switch and the indicator LED, and a piezoelectric beeper that will sound while the bottom float is low and the circuit hasn't shut the fuel off yet.

The high float switch circuit simply drives the red LED directly. The low float switch (SW2) drives Q1 and Q2. When the tank is full, SW2 will be open, and Q1 will be biased ON by current flowing through R3, R1 and R2. Q1's collector will be high, and C1 (the main timing capacitor) will charge quickly through R4 and D1, and remain fully charged.

When the fuel level dips below the bottom float, SW2 will close, and this will pull the left side of R3 up to the supply voltage, removing the bias from Q1, so Q1's collector will go low. D1 will now be reverse-biased and will no longer keep C1 charged, and C1 will slowly discharge through RT, which is a preset (screwdriver-adjustable) potentiometer called a trimmer or trimpot.

While C1 is charged, current through R5 into Q3's base keeps Q3 turned ON, which removes the bias from Q4, which stays OFF, so the relay coil is not activated.

Once the C1 voltage drops below about 1.5V, Q3 turns OFF, and Q4 turns ON. Current flows through the relay coil and Q4, and causes a 0.7V voltage drop across D3, raising the emitter voltages of Q3 and Q4. This biases Q3 further OFF and the circuit "snaps" into this state.

The relay contact must be wired so that when the relay is energised, the fuel supply will be cut off. This might involve using the normally closed contact on the relay.

R8 is a high-energy varistor to protect the circuit against overvoltage spikes on the incoming power supply.

The piezoelectric beeper BZ1 is driven by Q2, which turns ON when the fuel level goes below the low float switch, and is powered from Q4 collector, so it will not beep if the relay is activated (in that state, Q4 collector will be at about 2V which isn't enough to run the beeper).

All of the parts are available from Digikey. I've specified three 4-pin plug/socket pairs to make the design look tidy. I've specified vertical headers but right-angle ones are available too. I've specified two alternative beepers. The second one is 20% more expensive, but 20 dB louder. No components are critical. You can use a 1N4001 for D1 as well if you like, to shorten the parts list. I haven't specified the relay type, since you already have it. RT is a single-turn 100 kilohm trimmer. Connect to one end and the wiper. You'll need some kind of prototyping board - I recommend stripboard aka Veroboard.

How do you feel about the project now?
 

Attachments

  • Steamngn.gif
    Steamngn.gif
    13.2 KB · Views: 1,523
Top