Electronics guys- transistor/relay question

Status
Not open for further replies.
T

The_Eric

Joined
Mar 31, 2010
Messages
6,995
Location
Iowa
Hey guys, I'm helping my shop partner work on his Deere 326D skidder. It's been plagued with intermittent long cranking, a service code that cannot be read through the cockpit display and the whole reason why it's getting attention now is the cooling fan will not run, so it will overheat.

While cleaning terminals and checking things over, I decided to check resistance across the glow plug relay so I got my little boost pack out and hooked it up to the coil terminals. Well, the current draw was through the roof! My 18ga jumper wires got hot in a heartbeat. Obviously not normal. I looked up the relay which is VF7-11F11-Z19. Couldn't find the exact same relay, but it looks like coil resistance should be 70-90 ohms and coil current draw on a similar number was 166mA. My coil resistance was .4 ohms. So I'm guessing the relay is toast, but more importantly I'm concerned about the coil driver (transistor) in the ECM. Can I assume that it's toast? Schematics show the relay coil wiring to go straight to ground and and straight to the ECM...

I know that the number of the transistor will dictate it's load carrying ability, but if it was only meant to turn on a relay and it saw a short, could it survive?
 
I very much doubt the relay driver circuit would be designed as to not tolerate a short. It's just good engineering. You're sure you tested the right terminals?
 
Sometimes the relay will have a free wheeling diode built in, to take care of the inductive kick when the relay is de energized. A good designer will also include one in the driver circuit to protect it. If the relay has this diode, it is polarity sensitive when applying 12V. If you forward bias this diode you will essentially short your power source through it. Ultimately this can cause the diode to short, then ultimately open if the fuse current is exceeded.
This diode could have failed in your relay, if it has one. As Papa Bear mentioned above, the "typical" failure mode for a relay coil to fail is open circuit.
As far as your driver circuit, you describe a high side switched circuit. While not unheard of, typical semiconductor drivers are low side switched. See the datasheet links attached. Most driver circuits that I work with (aerospace) are fault tolerant and will provide the status of the load. I have seen some poorly designed drivers that failed due to a short circuit or somebody installing a relay not equipped with a diode.

http://datasheets.maximintegrated.com/en/ds/MAX4896.pdf

http://www.onsemi.com/pub_link/Collateral/NUD3112-D.PDF

I'd verify your diagram is correct and that you connected the relay right. The main concern would be if the ECU driver is good. You could test the ECU driver by using an equivalent resistor (resistance and wattage) or a test light (again not exceeding the capability of the driver) connected to where the relay coil would be and verify you get 12V when you should. If you check it open circuit, without a load, you can get a false indication of 12V due to leakage current, if it is a semiconductor driver.
 
Last edited:
Originally Posted By: meadows
Sure you were not putting power to the glow plugs?


That much I am positive of. The relay was out of the machine and on the bench.



As for making the correct connections, that I'm sure of too. There are 4 terminals- 85, 86, 87 and 30.

85 and 86 are the coil terminals and 87/30 are for the load.

JetStar- good call on the diode. The schematic on the relay case does show one. Of course I didn't notice it immediately and may have wired them backwards at first. The machine schematic shows 85 as negative and 86 as positive. Regardless, either way they're wired causes arcing and hot wires.

I also thought being high side switched was odd, and as you've noticed, basically everything I've seen in auto/light trucks has been low side switched... As an aside- why is that?

So, now that I can probably safely assume that the driver is toast (actually, I kinda already did) should I be able to see it? I peeled the ECM open to inspect for burns, corrosion or other signs of foul play but can find none. I REALLY hate to miss my diag on this, as the cheapest used ECM I have seen is about a grand! New is 2 grand!
 
The_Eric I also thought being high side switched was odd said:
Mainly it has to do with the level of base voltage (in the case of a transistor) or gate voltage (in the case of a mosfet) required to turn the device on. In the case of an N-Channel Mosfet in a 12V system, it would require >12V to turn it on and drive a high side switched load. It can be done, but not very convenient in a 12V system.

Those terminal numbers on the relay sound a lot like the industry standard fog light, fuel pump, etc... relays that we're all familiar with.

Note that these can be had with the diode oriented both ways.
http://www.delcity.net/store/Relay-with-Diode/p_73578.r_IS1018?&gclid=CJzD95KSxcoCFQ8zaQodGeEE1A
Click to expand...
 
You could test ECM and coil circuit by using a small bulb in place of the relay coil. See if bulb lights and goes out at appropriate times.

Do not plug the shorted relay back in, obviously you need a new relay.
 
Testing the driver outside of the machine isn't really practical. That would require powering up the ECM and fooling it into thinking the machine was hot enough to need the fan. About all we can do is reinstall it and run the machine to temp (with a replacement relay in place) and see if the fan kicks in.

Here are a couple of pics of said relay- it's a pretty heavy duty one. It has a 60 amp fuse in the circuit and relays with similar numbers are rated for 70 amps.



 
That relay is a very common one It is used in electric forklifts that I used to work on and in one of my friends old stern drive F&R shifting circuit. interesting. glow plugs take lots of current.
 
Well I just looked through the Sierra marine catalog and couldn't find that relay. I looked through OMC, Mercruiser, Mercury (outboard) and Volvo-Penta... A few relays, but none like that.

I wonder if we could rig up a heavier style relay- like one used for starting small engines used in riding mowers? A ECM driver should be able to handle that kind of current draw, shouldn't it?
 
Tom you may be right. To be honest, I'm not 100% sure if it needs a diode or not. I get the impression that the relay has been replaced before... I looked up the Deere number (AT322221) but couldn't find much in the way of info or pictures.
 
I did a quick search, and Digikey's page indicates the wrong datasheet.
http://www.digikey.com/product-detail/en/VF7-11F11/PB708-ND/813775

but it lists a sub:
http://www.digikey.com/product-detail/en/G8JR-1A7T-R-DC12/Z2716-ND/1618708

Mechanically it looks similar...
https://www.components.omron.com/components/web/pdflib.nsf/0/E6C4C0E4A0A85DDF85257201007DD648/$file/D20G8JR1005.pdf

I forget if Digikey still has min qty order, so here is Mouser's version:
http://www.mouser.com/ProductDetail/Omron/G8JR-1A7T-R-DC12/?qs=%2fha2pyFadujcoPGtr%2fVmXeXN1RVLa%2flRxVNepPPSQgGU3LHLi8EJNQ%3d%3d

Poking around on TE Connectivity's website I can't find a VF7 series. Makes me think they changed numbering schemes or somesuch. If I get a chance later I'll look again.

If you are concerned you can always tack in a diode across the wiring harness, and use a non-diode. Any 1N4000 series diode will work for something like this. Not sure on the application and if you want to mess with installing a diode (concerns with hacking the harness, difficulty in soldering, environment concerns, etc).

If you are concerned about the driver being kaput, try Dremeling around the case of the bad relay, and remove the guts. Find a 1W or so 12V bulb and tack it across the winding terminals. Install and see if the bulb turns on when it should.
 
Originally Posted By: The_Eric
Testing the driver outside of the machine isn't really practical. That would require powering up the ECM and fooling it into thinking the machine was hot enough to need the fan. About all we can do is reinstall it and run the machine to temp (with a replacement relay in place) and see if the fan kicks in.

No I was saying have the machine all put together (except for the relay); test light across the relay coil. Operate until the relay should come on and see if the light lights up.
 
Not much of an update... The machine is still sitting until we can get time enough to work on it. Work partner is repairing/replacing some wiring and working on souring a replacement PCM.

A note on that relay... He went out and bought a brand new OEM relay from Deere- and it looks EXACTLY like the one in the picture. I thought for sure it had been replaced at one time, but evidently not.
 
Status
Not open for further replies.

Similar threads

Don_M
Why relay placement is important
Replies
9
Views
1K
Don_M
A
A/C clutch fuse keeps blowing after high speed radiator fan kicks in.
2
Replies
20
Views
3K
Nukeman7
buck91
  • Locked
Relay harness fuse size
Replies
13
Views
4K
buck91
G
  • Locked
Car audio gurus chime in
2 3
Replies
43
Views
5K
grampi
S
  • Locked
HVAC Compressor Quit
2 3
Replies
51
Views
8K
dnewton3
Back
Top Bottom