Maker Pro
Maker Pro

12v timer for pump

Bradbrad

Jul 4, 2013
1
Joined
Jul 4, 2013
Messages
1
Hi everyone. This is my first post and I'm no electronics expert. Herein lies my problem and would welcome any assistance....

I want to run a small 12v oil pump for approx 15secs every 10mins.
I would like the pulse and the pause to be adjustable/variable so I can fine tune things.
I have located this item from my local electronics store : MK111 Adjustable Timer with Relay Output
http://www.maplin.co.uk/mk111-adjustable-timer-with-relay-output-25224

The problem is the pulse and pause setting are not suitable for my needs. Can anyone tell me what component I would need to change in this circuitry and where I could source such a part?

The oil pump is rated at DC12v; 3.6W; 0.3A

Thanks to anyone who can help me with my dilemma

Brad
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Hi Brad and welcome to the electronics point forums :)

You can do this with a 555-based oscillator driving a transistor or MOSFET.

Have a look at the basic 555 oscillator. It's the first diagram on this page:

http://www.electronics-tutorials.ws/waveforms/555_oscillator.html

You would need to replace the two resistors with potentiometers, to make the high and low periods adjustable. R1 controls the longer period - the ten minute delay. R2 controls the shorter period - the 15 second run time. (Actually, R2 also affects the delay between run periods, though it doesn't contribute much compared to R1 because it's a much lower value.)

Ten minutes is quite a long time for a 555 oscillator. You would need to use a high capacitor value for C1. Assuming you use a 1 megohm potentiometer for R1 (the highest value that's widely available), and you use a 1000 uF (microfarad) capacitor for C1, the maximum interval will be about 12 minutes. Using a 50 kilohm potentiometer for R2 will give a maximum pump run time of about 35 seconds.

C1, the 1000 uF capacitor, will probably need to be an aluminium electrolytic. These are not very accurate, and their values vary somewhat with temperature and age. Here's one from Nichicon, one of several high-quality Japanese brands:

http://www.digikey.com/product-detail/en/UBT1E102MHD1TO/493-4500-1-ND/2650340

Here are some more links to Digikey part pages for you to look at:

1M potentiometer, slotted shaft: http://www.digikey.com/product-detail/en/51AAD-B24-A25L/51AAD-B24-A25L-ND/3828851
1M potentiometer, four-turn (easier to adjust accurately): http://www.digikey.com/product-detail/en/3856A-282-105AL/3856A-282-105AL-ND/3828850
50k potentiometer: http://www.digikey.com/product-detail/en/93R1A-R22-A18L/93R1A-R22-A18L-ND/2564723

There are many possibilities.You can check out other potentiometer options using Digikey's selection table: http://www.digikey.com/product-sear...sistors/rotary-potentiometers-linear/?stock=1

The only really important parameters are the resistance, and the mechanical style (mounting style, shaft style). I would avoid "conductive plastic" and wirewound ones; carbon is OK and cermet is the best quality material.

Digikey have potentiometers with values of 2.5 megohms and 5 megohms; you could use one of these, to allow you to use a lower-value capacitor. Reduce the capacitor value from 1000 uF by the same factor that you increase the potentiometer resistance by. You'll also need to increase the 50k potentiometer in line with the 1M potentiometer.

TLC555 (improved version of the 555 timer): http://www.digikey.com/product-detail/en/TLC555CP/296-1857-5-ND/277502

NTD2955 MOSFET: http://www.digikey.com/product-detail/en/NTD2955-1G/NTD2955-1GOS-ND/1484753

That MOSFET will be suitable for your application. Don't worry that it's rated for 12A; it's the cheapest suitable MOSFET on the list!

Here's how to connect it.

Connect the MOSFET's source to the +12V supply, which also powers the TLC555.
Connect the MOSFET's gate through a 22 ohm resistor to pin 3 of the TLC555.
Connect the MOSFET's drain to the positive terminal of the pump.
Connect the negative return from the pump to the circuit's negative rail (called the "0V rail").
Connect a diode across the pump, i.e. from the MOSFET's drain to the circuit's 0V rail, with its cathode (the stripe end) to the MOSFET's drain. This protects the MOSFET against inductive kickback (aka back-EMF) from the pump, which could damage it. A suitable diode would be the 1N4001:
http://www.digikey.com/product-detail/en/1N4001/1N4001FSCT-ND/1532742

You'll need to build the circuit up on stripboard, also called veroboard. Google stripboard assembly tutorial for information.

Any other questions please feel free to ask.
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Just be very careful with a 555 running at extended times. Input leakage current, noise, capacitor leakage, etc. all conspire to make your circuit unreliable.

If you're lucky, this will just vary the delay, if you're not, the oscillator can stop.

There's a thread here somewhere where an example is given showing how the charge current falls into the range of leakage currents for the capacitor.

A slightly more complex (but more stable) answer is to have a 555 having a 15 second ON time and a 45 second OFF time. This is fed to a 4017 which does a divide by 10. One output of the 4017 and the 555's output are ANDed together to switch the relay.

Whilst 45 seconds is getting long for a 555, it's not close to where you'll have real problems. So it will be much more tolerant to temperature changes and aging of components.

An even simpler solution would involve (say) a picaxe doing the timing. However that requires a bit of programming.

edit: you could also have a 45 sec ON, 15 sec OFF for the 555 which would work with the conventional circuits.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
A slightly more complex (but more stable) answer is to have a 555 having a 15 second ON time and a 45 second OFF time. This is fed to a 4017 which does a divide by 10. One output of the 4017 and the 555's output are ANDed together to switch the relay.
Hey, that's a clever idea!

Yes, ten minutes is a long time for a 555 oscillator.

I tried to find low-leakage electrolytics. A couple of Japanese manufacturers - Rubycon and Nichicon, I think - have low-leakage series, but they weren't available through Digikey. At least, not 1000 uF. So I suggested one that's rated for a very long lifetime at high temperature. I figured that since leakage increases with temperature, a cap that's rated for high temperature and long life would have stable leakage characteristics (as well as being generally high-quality). Not sure if that's a reasonable assumption though.

It's a bit irritating that the period of the 555 oscillator can't be set independently of the ON time. You would have to use a monostable to allow the interval and the run time to be set independently. A 556 with a 4017 would be a good solution for this application.

Actually, even better would be a 556 with a 4024. I'll draw up a schematic :)
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
attachment.php


U1 contains two timer circuits. The left hand circuit is used as an oscillator. Its frequency is controlled by VR1, R1, R2 and C1. With the values given, the period between output pulses is adjustable from 3.36 seconds (VR1 fully counter-clockwise) to 6.65 seconds (VR1 fully clockwise).

This pulse stream is fed into the clock input of U2, a CD4024 seven-stage ripple-carry binary counter. The output from the last stage, Q7, toggles with a period 128 times that of the clock, i.e. between 430 seconds (7 minutes, 10 seconds) and 850 seconds (14 minutes, 10 seconds).

All timing figures are approximate; all components have tolerances (C1 has a basic tolerance of +/- 5%). They will also vary somewhat with temperature, and slightly with age and battery voltage variations.

The generated pulse stream is used to illuminate LED1 briefly. You can calculate the run interval by timing the interval between flashes and multiplying by 128.

On every falling edge (transition from high to low) on U2's Q7 output, a negative pulse is compled through C3 into the trigger input of the right hand half of U2, which is connected as a monostable (a pulse generator). It generates a low-going pulse on its output, which turns MOSFET Q1 ON, feeding current into the load.

The period of the monostable is determined by VR2, R4 and C2, and is adjustable from 5.4 seconds (VR1 fully counter-clockwise) to 22 seconds (VR2 fully clockwise).

The circuit operates from any supply voltage in the range 6V to 14.0V. Voltages higher than 15V can damage U1 and Q1. C5 provides decoupling for U1 and U2 and should be connected as close as possible to pins 14 and 7 of U2.

Load currents up to several amps can be supplied easily, provided that the battery can supply enough current without its voltage dropping significantly.

I have provided a components list. This is not a complete bill of materials. Other items required are connectors for the battery and the load, and a piece of prototyping board such as stripboard. I have not drawn up a stripboard layout; you'll need to work that out yourself. Good luck!

Edit: The NTD3055 is not in a TO-220 package as stated on the parts list. It's in an IPAK package.
Edit2: I have given two Digikey part numbers for the trimpots. The 12-turn ones can be set more accurately than the single-turn ones.
 

Attachments

  • BRADBRAD.001.GIF
    BRADBRAD.001.GIF
    16 KB · Views: 248
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
The Maplin module you linked to is a good design. Ten minutes is a long time for a 555-based oscillator but it can work if you use a good-quality timing capacitor. That's C3 and you need to increase it to about 1000 uF. Don't expose it to high temperatures; this will increase the leakage in the capacitor.

The way that circuit is designed, leakage in C3 will not cause the oscillator to stop. It will just reduce the "pause" interval.

(Leakage, in this sense, is electrical current leakage. It doesn't relate to physical leakage of the electrolyte out of the capacitor.)
 
Top