Solution for Hard Starting When Warm

[Carimbo]

This seems to be a common problem: Hard starting when warm, i.e. after sitting 30min. - 1-1/2 hrs. The glow system takes its cue from the temp sensor in the thermostat housing. Problem is if the cylinder temp has cooled too much for easy combustion w/o glowplug, but the temp sensor is still reading high enough temp that tells the glow system to not energize the glow system.

I had been fooling the system in these situations to energize the GPs by disconnecting one of the butt connectors at the temp sensor, starting the car, running back to the front of the car to quickly reconnect the sensor, shut the hood and off we go. Works but inconvenient.

Yesterday I wired in a simple SPST toggle switch to cut the signal in the same circuit, inside the car. The autoglow box is under the passenger seat in my car. Removed the seat (4 easily accessible bolts, light seat), unplugged the connection to the box, cut the green wire (blue one would also work), spliced in some lengths of 16ga. to a SPST toggle switch. Have not yet decided where to mount the switch; for now it is alongside the console. The leads are long enough to reach to the dash/instrument panel if I want. Reconnected/reinstalled everything and had a chance to test it yesterday while running errands.

Car didn't want to start easily after 45 mins. shutdown. Instead of grinding on the starter, I flipped the switch to break the connection to the temp sensor, key to on position, let the GPs energize for about 2-3 secs. and then it fired immediately. Flipped the toggle back and we were off.

In the future may adopt a more elegant solution but for now this works, and is much more convenient, and hopefully will help conserve the starter and not fry the GPs.

[asavage]

Yeah, I have the warm-start problem too. I understand what you're doing. Have you tested how long the GPC would keep the GPs lit if you left the switch on?

I'm thinking that, rather than toggle the switch twice, I'd put a momentary contact switch that energizes the GP relay instead. That way, the GPs are lit only when I push the button (and/or when the GPC energizes the relay). The GP system can't have the GPs on too long, and the GPC still functions normally regardless of the whether I'm on the switch or not. Can also add add'l glow time for cold starts if deemed necessary, rather than cycling the Ign. switch.

The setup I describe is what a lot of Ford 6.9l folks do, and is generally what I've seen engine transplanters use to simplify wiring for the lazy. It's also likely what Steve's wagon had set up. The only reasons I haven't added this yet is that a) I'm very lazy and my LD28s are not my daily drivers, and b) I want to also wire a dash indicator connected to the GP bus bar that lets me know when there's actually juice to the GPs.

[philip]

This seems to be a common problem: Hard starting when warm, i.e. after sitting 30min. - 1-1/2 hrs. The glow system takes its cue from the temp sensor in the thermostat housing. Problem is if the cylinder temp has cooled too much for easy combustion w/o glowplug, but the temp sensor is still reading high enough temp that tells the glow system to not energize the glow system.

SNIP

Car didn't want to start easily after 45 mins. shutdown. Instead of grinding on the starter, I flipped the switch to break the connection to the temp sensor, key to on position, let the GPs energize for about 2-3 secs. and then it fired immediately. Flipped the toggle back and we were off.

I'm not familiar with LD28 specifics so I'll generalize from the SD22.

It seems your sensor gains resistance as it cools so ... it seems you should find the right resistor to add into the circuit so that the threshold for glow plug operation occurs at a higher temperature. No? I'll be the first to admit my eletronics knowledge is quite limited.

Here is the resistance curve for the SD22 which uses a dedicated temperature sensor in the AutoAfterGlow system.

Temperature Resistance
14º F 7.0-11.4 kΩ
68º F 2.1-2.9 kΩ
122º F 0.6-1.0 kΩ

[Carimbo]

Have you tested how long the GPC would keep the GPs lit if you left the switch on?

I use the switch "off" position to cut the temp signal to the GPC, fooling it to think the engine is cold cold (infinite resistance). Have not actually tested how long the GPC will energize the GPs in this mode but presume it would be the same as a normal cold start on a very cold morning.

I'm thinking that, rather than toggle the switch twice, I'd put a momentary contact switch that energizes the GP relay instead.

Well I did go looking for a momentary contact switch for this application (actually a momentary break-contact switch) hard to find (now that I think about it, probably a salvaged refrigerator door light switch would work), so I wired in the simple toggle switch. Not too much of a hassle to switch it to off, start the car, switch it back on. Currently I'm happy to avoid diving under the hood to disconnect/reconnect wires in the dark.

Your idea is certainly more surgical and to the point, controlling only the part of the system we want to affect. One thing I did notice when cutting/reapplying the temp sensor signal-- the airbox throttle butterfly snaps open/shut according to the existance of the temp signal. Bet it affects the cold start timing auto-advance as well.

[asavage]

One thing I did notice when cutting/reapplying the temp sensor signal-- the airbox throttle butterfly snaps open/shut according to the existance of the temp signal.

The EGR flow is disabled when the engine is cold. Unlike any other engine I've ever seen, the LD28 EGR is active at idle when it's warmed up. The EGR valve opens, and the throttle butterfly closes, when activated. At idle. I didn't believe it until the Crown diesel book pointed it out in the LD28 troubleshooting section, and the FSM confirms this operation, and I then stuck my finger on the EGR diaphragm: open at idle when engine is warmed up.

This is all controlled by that Black Box I pulled off the driver's side footwell at the JY -- the one I initially thought was the GPC. Both the GPC and the EGR controller use the same temp sensor for input.

Bet it affects the cold start timing auto-advance as well.

No, the CSD is entirely thermal, a wax pellet surrounded by coolant. It has no external influences, only the coolant temperature. It props the idle speed and also advances the timing up to some RPM.

[goglio704]

My mod to the GP system was to put a momentary push button in to close 12 volt + to terminal 6 on the GP controller. This will energize the glow plugs through the dropping resistor. From the factory the plugs are energized at reduced voltage everytime you crank the engine regardless of temp. Don't believe me - hook a light in parallel with the plugs and you'll see. Anyway instead of cranking the engine for 5 to 10 seconds when it is in the dreaded warm range I energize the plugs at reduced voltage for 5 to 10 and it fires right off. What I like about this mod is that it should be hard to fry the plugs at reduced voltage. It is not as fast, but it is safer. The plugs with the requlating coil (NGK) are real expensive so I try to be kind to them. I also tried disconnecting the wire from the ignition switch so I wouldn't energize the plugs and suck the battery down during cranking. I ran out of cold weather before I decided if I liked it better that way or not.

P.S. I agree with Philip about the resistance curve being tuneable by adding resistors. In fact if you look hard I suggested it in a post a while back. (Page2 post2 of "glow plug circuit problemo")

[davehoos]

the coolant sensors come in a variety of calibrations.normally the tolerance is colour coded.i find the white sencor give a HOTTER reading,i use these in petrol models to lean the fuel off earlier.if it is hard to start it sounds more like the water temp is not hot enough to heat the metal and oil.

other diesel makers put a simple switch in the thermostat outlet to test for hot starting.you unplug it if the car is used for stop-start operation...the most advance controler was the isuzu MARK I[gemini] Diesel-KIKI .the later fast glow systems are based loosely on it.these had what called smooth idle control for warm up and start -stop aplication.
the old nissan system simply conected to the starter using the low voltage of cranking to reduce the overheating of the plugs.

[asavage]

My mod to the GP system was to put a momentary push button in to close 12 volt + to terminal 6 on the GP controller. This will energize the glow plugs through the dropping resistor. From the factory the plugs are energized at reduced voltage everytime you crank the engine regardless of temp.

(regarding GPC terminal 6) A terminal connected to the "START" position of the ignition switch (When the ignition key is returned from "START" to "ON", after-flow operation begins.)

However, what you said seems to be true with my '82 Wagon: Turning the key to Start engages the Afterglow relay and I measure voltage at the GP bus. And when I let the key spring back to Run, the Afterglow timer is initiated -- the Afterglow relay continues to run for a while.

For those of us that have the automatic trans, here's another trick: put the shift selector in any gear except Park or Neutral, then hold the key to Start. The starter will not operate (due to the transmission neutral safety switch) but on my '82, the Afterglow relay is engaged anyway. No wiring additions needed

P.S. I agree with Philip about the resistance curve being tuneable by adding resistors. In fact if you look hard I suggested it in a post a while back. (Page2 post2 of "glow plug circuit problemo")

Problem is, the resistance curve of the sensor is non-linear. Adding a series fixed resistor only affects the lower part of the curve.

[goglio704]

Kinda makes me wish the manual had a clutch interlock that would function like the Neutral and Park interlock on the automatic. I'm not normally a fan of being supervised by my machines, but at least this would have a side benefit. I think the afterglow time after cranking diminishes substantially or even disappears as the engine gets warmer. I also have a light wired in on the car with the mod, but I don't drive it much because Mom has it.

I hadn't looked at the tuning resistor idea lately, but IIRC the resistance grows with colder temps. Unplugging it produces infinite resistance and increases the time. My thought was to add a fixed resistor in series that would produce about a 30 degree shift in how the warm range is interpreted. At the time I thought it would be an insignificant change in the colder ranges because the resistance is a lot higher when cold, but I've been wrong before. To put it another way, I thought the non linear aspect was actually a help.