PDA

View Full Version : Possibly a nieve question about diodes



cygnusv
06-12-13, 16:44
Our fresh water pump is a 42PSI 5.2GPH Johnson. When the pump started failing on a VERY regular basis because of the pressure switch I asked for and received very good advice from this forum. As a result of this advice I fitted an external Whale pressure switch and a 24V relay. When the relay failed I asked for more advice and learned about back EMF and the benefit of fitting a diode across the relay. We now get trouble free operation for about 6 months when we again have to replace the relay.

My question is: Would there be any advantage in soldering 2 diodes together and effectively using 2 of them rather than the single one we've currently using now?

The relay is a 5 pin 24V 30-40amp automobile type

The diodes are 1N4007 - M1C. This from the tiny printing on the diode

I believe that the pump is rated at about 8 amps

Thanks in advance, Stu

lw395
06-12-13, 17:15
If the failure is being caused by arcing, it might help to replace the diode with some sort of zener diode, which will conduct in either direction if more than say 35V are across it.
Perhaps something like:
http://uk.rs-online.com/web/p/products/486-4656/

Alternatively replace the relay with a suitable power FET?

superheat6k
06-12-13, 17:19
I would wire an automotive coil condenser (capacitor) across the pump connections at the relay. This should suppress the arc that is likely wrecking your relay each time it opens (turns off). eBay Condenser in car parts or try a motor factors.

Alex_Blackwood
06-12-13, 17:36
Our fresh water pump is a 42PSI 5.2GPH Johnson. When the pump started failing on a VERY regular basis because of the pressure switch I asked for and received very good advice from this forum. As a result of this advice I fitted an external Whale pressure switch and a 24V relay. When the relay failed I asked for more advice and learned about back EMF and the benefit of fitting a diode across the relay. We now get trouble free operation for about 6 months when we again have to replace the relay.

My question is: Would there be any advantage in soldering 2 diodes together and effectively using 2 of them rather than the single one we've currently using now?

The relay is a 5 pin 24V 30-40amp automobile type

The diodes are 1N4007 - M1C. This from the tiny printing on the diode

I believe that the pump is rated at about 8 amps

Thanks in advance, Stu

Stu, First of all. Is your system 24volt? Is the relay similar to the one on e-bay?
http://www.ebay.co.uk/itm/CAR-BIKE-24V-40-AMP-5-PIN-CHANGEOVER-RELAY-SWICH-X-5-/400165316805
I assume that the answer to my first question is yes. If so and the answer to the second question is also yes. Then the relay contacts are rated at 14 volts. That could be why the relay is failing. The higher voltage will reduce the current rating of the contacts. Also the diode you quote is only rated at 1 amp.
You don't say where you fitted the diode. Is it across the contacts or the coil? I assume the coil. However with a rating of only 1 amp I would have thought that a bit light to cope with inrush current or back EMF puls from the coil. Also you don't say how the relay fails. Is it the coil that fails or the contacts that burn out.
Why did you fit a relay in the first place? Was it because the pressure switch contacts were not rated at a high enough current for the pump?
Sorry not to be of more help but from the given info. there could be a variety of reasons for the failure.

cygnusv
06-12-13, 18:09
Stu, First of all. Is your system 24volt? Is the relay similar to the one on e-bay?
http://www.ebay.co.uk/itm/CAR-BIKE-24V-40-AMP-5-PIN-CHANGEOVER-RELAY-SWICH-X-5-/400165316805
I assume that the answer to my first question is yes. If so and the answer to the second question is also yes. Then the relay contacts are rated at 14 volts. That could be why the relay is failing. The higher voltage will reduce the current rating of the contacts. Also the diode you quote is only rated at 1 amp.
You don't say where you fitted the diode. Is it across the contacts or the coil? I assume the coil. However with a rating of only 1 amp I would have thought that a bit light to cope with inrush current or back EMF puls from the coil. Also you don't say how the relay fails. Is it the coil that fails or the contacts that burn out.
Why did you fit a relay in the first place? Was it because the pressure switch contacts were not rated at a high enough current for the pump?
Sorry not to be of more help but from the given info. there could be a variety of reasons for the failure.

Hi Alex
Yes we are 24 volts and the relay I'm using is of the type shown I think. I'm a bit in the dark. The relays I've used have all been 24 volt ones ( Car Truck 24V 30/40 AMP CHANGEOVER RELAY SWITCH + Wiring Harness with Diode
Item Number 140742323825 ) Are these relay contacts rated at 14 volts? I think that the contacts are burning out because first signs are when the pump gets stuck 'ON' and the PRV on the calorifier blows off.

The diode is looking iffy if it's only rated at 1 amp. The pump operates at around 5-8 amps I think. Is there a straightforward replacement for the weak kneed diode I'm using?

The reason we're doing this is because, although this Johnson is a great pump, the pressure switch failed at least 6 times in the first year. At the time the pump was a pig to get to, so repair / replacement was a pig. We've since moved it to somewhere more civilised. At least with the way we are dealing with the problem we are getting acceptable service.


superheat6k

Thanks for your input too. Are condensers voltage sensitive? I ask because we're 24 volts. Also, how would the condenser discharge/ (worried about heavy duty sparks!)


lw395

Would a Zenner diode directly replace the small diode I'm using and are they polarity sensitive? What is a suitable power FET?

Hope you all don't mind me replying in this fashion, but I do appreciate you taking the time to reply and am interested in the various alternative solutions.

I look forward to your much appreciated replies, thanks, Stu

superheat6k
06-12-13, 18:46
The induced EMF being discharged when the contact inside the relay opens probably spikes at several thousand volts, its origin being the pump motor windings and is caused by the collapsing magnetic field that was driving the pump motor, hence the spark / arc. From your comment that this manifests with the pump stuck on, eventually the relay contact is welding itself closed (on), and this because the arcing destroys the conducting area of the relay contact surface, in turn all the current flows through an ever diminishing clean contact area. Eventually the current flow at this point meets a higher than normal resistance, so heat builds up, to a level where the contact surfaces melt and fuse together (welds).

So the operating voltage as far as the condenser is concerned is immaterial as to 12 or 24 vdc. The condenser simply works by temporarily absorbing this high energy / voltage spike, preventing the arc.

This is exactly the same as its function on a car's coil / points ignition system.

It can either discharge harmlessly back into the circuit next time it closes, or you could fit a ballast resistor which allows the condenser to discharge. Something like a 1k ohm 1/2 watt resistor (an expert will soon be along to correct me if I am wrong here), wired in parallel with the condenser / capacitor.

I am not sure how wiring a diode for this problem will help at all, but again I am sure someone can explain this. As I understand it a simple condenser / ballast resistor pairing is far easier to set up than Zener diode or FETs.

cygnusv
06-12-13, 19:04
The induced EMF being discharged when the contact inside the relay opens probably spikes at several thousand volts, its origin being the pump motor windings and is caused by the collapsing magnetic field that was driving the pump motor, hence the spark / arc. From your comment that this manifests with the pump stuck on, eventually the relay contact is welding itself closed (on), and this because the arcing destroys the conducting area of the relay contact surface, in turn all the current flows through an ever diminishing clean contact area. Eventually the current flow at this point meets a higher than normal resistance, so heat builds up, to a level where the contact surfaces melt and fuse together (welds).

So the operating voltage as far as the condenser is concerned is immaterial as to 12 or 24 vdc. The condenser simply works by temporarily absorbing this high energy / voltage spike, preventing the arc.

This is exactly the same as its function on a car's coil / points ignition system.

It can either discharge harmlessly back into the circuit next time it closes, or you could fit a ballast resistor which allows the condenser to discharge. Something like a 1k ohm 1/2 watt resistor (an expert will soon be along to correct me if I am wrong here), wired in parallel with the condenser / capacitor.

I am not sure how wiring a diode for this problem will help at all, but again I am sure someone can explain this. As I understand it a simple condenser / ballast resistor pairing is far easier to set up than Zener diode or FETs.

It sounds very feasible based on what you've said. I'll hang on before rushing out to get the bits in case someone can confirm (only based on your bracketed comment) and maybe give me an idea of suitable suppliers part numbers. I must say that it's going to be great to nail this once and for all!

Much appreciated Superheat6K and thanks, Stu

lw395
06-12-13, 19:48
1) the diode will be rated at 1A continuous but many times that for a short surge.
2) A condenser, AKA capacitor will help, but less so in the case of a motor than an ignition coil. The motor has physical inertia as well as inductance, when you open circuit the motor, it looks like a dynamo generating a current.
3) The diode gives this current a path and keeps the volts down. IF the diode is across the contacts, it forces the current to travel through the fuse, wiring, battery etc, Which may have a lot of inductance. You can get a better effect by putting a zener or transorb or other arrestor across the motor, so the current from the spinning motor has a short path that keeps the volts down to below that needed for a spark. Obviously this surge arrestor must not conduct when the motor has normal power applied to it. So it needs to be good for more than 28V, plus a margin for ripple. You could usefully have the normal diode across the contacts as well to pass the bulk of the energy back to the battery. Not to save the energy but to avoid it overheating the suppressor.
4) It is the way of things that electrical reality is less definite than people's opinions, as small changes in circumstance can have big implications. I'm not saying anyone is wrong!
5) You may find a better or different brand of relay lasts four times as long and it's no longer an issue...

6) A MOSFET like
http://uk.rs-online.com/web/p/mosfet-transistors/6715127/
might be part of a solid state solution, but buying a pre-built module might be more appropriate?
I'll have a look around if I get chance....

nigelmercier
06-12-13, 20:14
The relays I've used have all been [http://www.ebay.co.uk/itm/140742323825] ...
Just to be clear, the diode on this relay is across the coil. This is to protect only the input of the relay, in this case the float switch. A second similar diode could be used across the contacts if they are burning out. Wire it so the end with the stripe is connected to the contact that is at +24V (with the relay open).

Putting a condenser (non-technical name for a capacitor) across the contacts will cause more harm than good. Its function in an ignition system is to resonate with the coil and extend the spark!

Alex_Blackwood
08-12-13, 11:29
Perhaps some of these links may be of use. I would be inclined to have a word with the DVE people.

http://www.icpltd.co.uk/pdf/simplavolt.pdf

http://www.suregripcontrols.com/install_suppress.htm

http://www.directvehicleelectrics.co.uk/products/Relays_6_volt_12_volt_and_24_volt/Durite_Make_and_Break_Relays_24_volt/Durite_mini_make_and_break_30_amp_relay_with_diode _24_volt_072776_A_type_termination.html

Just a couple of questions. You have by fitting a relay obviously transfered the original problem from the pressure switch to the relay. What is the current rating of your pressure switch? If high enough it may be possible to dispense with the relay and fit a suppressor direct to the switch.
Also Do you need such a powerful pump for your system? perhaps it would help if a smaller pump was fitted.
Having said all that. As you have the system set up I would be inclined to look at the DVE relays with diodes.
Good luck!
Alex.

pampas
08-12-13, 14:56
Came across this problem on several boats.
1 +tv to relay ( motor circuit) output to motor then to -tv.
2. entirely seperate +tv to coils relay Pin, output from relay to pressure s/w thence to a SEPERATE earth point. I suspect the motor stops dead when the pressure switch opens producing a very large transient, I would be inclined to suppress at the closest point to the motor with a capacitor with a voltage rating around 50 volts minimum.
Trust this makes sence.
Gordon

nigelmercier
08-12-13, 15:17
... I would be inclined to suppress at the closest point to the motor with a capacitor
See my post above.

Marsupial
08-12-13, 16:42
Putting a condenser (non-technical name for a capacitor) across the contacts will cause more harm than good. Its function in an ignition system is to resonate with the coil and extend the spark!

Nigel, the capacitor in an old style coil ignition system has two functions, it directs back emf into the ignition coil to further boost the induced voltage AND it prevents the contacts from burning, it could be exactly what the OP needs but with a motor in the equation I would be very wary about using a relay at all.

nigelmercier
08-12-13, 18:55
Nigel, the capacitor in an old style coil ignition system has two functions, it directs back emf into the ignition coil to further boost the induced voltage AND it prevents the contacts from burning, it could be exactly what the OP needs but with a motor in the equation I would be very wary about using a relay at all.

It prevents contact burn by directing energy back into the coil. Are you suggesting that the pump motor would benefit in some way from being part of a resonant circuit?

The correct component to use is a diode, not a capacitor. It can either be across the relay contacts 87 and 30, or across the motor coil.

Having a motor in the equation makes no difference, it doesn't know it's being switched by a relay.

Marsupial
08-12-13, 20:16
It prevents contact burn by directing energy back into the coil. Are you suggesting that the pump motor would benefit in some way from being part of a resonant circuit?

The correct component to use is a diode, not a capacitor. It can either be across the relay contacts 87 and 30, or across the motor coil.

Having a motor in the equation makes no difference, it doesn't know it's being switched by a relay.

No Nigel I was correcting your description of an ignition circuit where the capacitor does benefit the circuit and prevent contact burn.

lw395
08-12-13, 20:54
A capacitor across switch contacts in an inductive circuit can reduce contact arcing by controlling the dI/dt.
It can be useful to limit the resonance by putting a resistor in series with the capacitor.
The inverse diode across the motor will provide a path for the inductive voltage from interrupting the motor current, but it does not provide a path for the back emf generated by the spinning motor.
For that, you either need a diode across the switch contacts to squirt it back to the battery, or a zener (etc) across the motor to clamp it to a bit over supply volts.
Or contacts that are up to the job of course!

nigelmercier
08-12-13, 23:25
For that, you either need a diode across the switch contacts ...
I think it would work across the coil, but this would be my suggestion also.


No Nigel I was correcting your description of an ignition circuit where the capacitor does benefit the circuit and prevent contact burn.
No correction required, I didn't state the converse.

William_H
09-12-13, 01:05
As said I think OP may have fitted the diode across the relay coil. This will only protect the pressure switch contacts.it is the mortor Back EMF that is burning the relay contacts.
As said I would be inclined to fit a zener diode across the relay contacts or across the motor supply. The zener diode should have a breakdown voltage around 40 volts. I would suggest that it have a power rating of 1w rather than the more common .3w rating. The zener didoe is connected with the band or point of the arrow to positive supply.
A zener diode is like a normal silicon diode which conducts current once the band gap voltage has been exceeded usually .7 volt. However in the reverse or blocking direction it has a breakdown voltage. It conducts current once that voltage is exceeded. You choose a voltage rating from 3v to 75 volts or more. Whereas a silicon diode has a breakdown voltagee usually 400v or more once this exceeded the diode is dead. In the zener diode in conducts current but recovers once voltage is reduced. ie designed to breakdown.
One exaple is 1N4752 33v 1w about 50 p cost. good luck olewill

Marsupial
09-12-13, 10:16
wade through this lot

http://www.omron.com/ecb/products/pdf/auto_precautions.pdf

enjoy

nigelmercier
09-12-13, 10:36
wade through this lot...

Or just Section (6).

Alex_Blackwood
09-12-13, 17:38
Perhaps some of these links may be of use. I would be inclined to have a word with the DVE people.

http://www.icpltd.co.uk/pdf/simplavolt.pdf

http://www.suregripcontrols.com/install_suppress.htm

http://www.directvehicleelectrics.co.uk/products/Relays_6_volt_12_volt_and_24_volt/Durite_Make_and_Break_Relays_24_volt/Durite_mini_make_and_break_30_amp_relay_with_diode _24_volt_072776_A_type_termination.html

Just a couple of questions. You have by fitting a relay obviously transfered the original problem from the pressure switch to the relay. What is the current rating of your pressure switch? If high enough it may be possible to dispense with the relay and fit a suppressor direct to the switch.
Also Do you need such a powerful pump for your system? perhaps it would help if a smaller pump was fitted.
Having said all that. As you have the system set up I would be inclined to look at the DVE relays with diodes.
Good luck!
Alex.

Sorry. I must correct my glaring error. The relays in the direct vehicle electrics link have the diodes across the coil. Whilst fine for protecting the pressure switch contacts, the OP also needs a diode across the motor input as has already been suggested. My fault for trying to rush on a Sunday morning!