Hard starts: No heat, no volts, no glow?

[Old Smokey]

Hey everyone,

I've read through all the other threads on hard starting but am more confused by the numerous potential causes than anything else.

I'm experiencing difficult, shaky starts with multiple stalls that smooth out in a minute or two. According to other posts, this would be symptomatic of a faulty GP system.

I have a question that came about following a multimeter test on my battery and GP system per Al's instruction:

Quick check: Check battery voltage at the battery with GP system energized. Is it greater than 11v? If so, connect voltmeter negative lead to No. 1 GP terminal, and positive lead to battery positive, energize GP system. This measures the voltage drop across all the GP system's load connectors and the relay. Ideally, you'd like 0.1v measured at the voltmeter; real-world, I'd probably be OK with anything up to a half-volt. If you're measuring two or three volts, you've found your problem (in a general sense): you are "dropping" voltage under load.

Battery reads 12v with GP system energized or not. Upon testing voltage between the positive battery terminal and the No. 1 GP terminal with system energized, I get a reading of about 2.5v until the GP relay switches again. At that point, the voltage shoots back up to 12v. I'm confused about the wording of Al's test. I can't tell if I'm supposed to look for a slight voltage drop FROM 12v or just a low voltage reading. Can someone please make sense of what I've described? I'd be happy to clarify further.

I suppose it's worth mentioning that there's a faded sticker on the side of this Autolife battery reading May 2004.

Also, should I be able to feel any warmth directly around the glow plug socket while/after they have been engaged? I feel nothing. Cold as ice.

Thanks,

Andy

[asavage]

Quick check: Check battery voltage at the battery with GP system energized. Is it greater than 11v? If so, connect voltmeter negative lead to No. 1 GP terminal, and positive lead to battery positive, energize GP system. This measures the voltage drop across all the GP system's load connectors and the relay. Ideally, you'd like 0.1v measured at the voltmeter; real-world, I'd probably be OK with anything up to a half-volt. If you're measuring two or three volts, you've found your problem (in a general sense): you are "dropping" voltage under load.

Battery reads 12v with GP system energized or not.

That's unusual. The load of the GPs (inrush of about 125A, steady drain of around 40-60A after about ten seconds; that kind of load should budge the battery downward while the GPs are doing their thing.

Generically, a fully-charged, fresh lead-acid automotive battery has a resting voltage of 12.6v.

Upon testing voltage between the positive battery terminal and the No. 1 GP terminal with system energized, I get a reading of about 2.5v . . .

Imagine that the all the wiring between the battery and the No. 1 GP is huge -- like as thick as your wrist. With the GP relay engaged, and this huge ideal wire with no resistance, the wire between the battery positive terminal and the end of the GP chain (which is No. 1 GP on the SD) would "use" no volts, no matter how much current the GPs consume. The wire and such can carry as much current as can be provided by the battery, and the same voltage (pressure) would be at the battery and the No. 1 GP.

In the real world, the wiring is always undersized, and multiple connectors each "eat" a little of the voltage -- they do not have zero resistance to current flow. On the 720, the primary culprit is the fusible link (if it still exists) at the battery positive terminal, but there are other connectors in series with No. 1 GP as well.

By connecting the VM as I described, you are measuring how much voltage the whole GP supply chain is "eating" -- voltage that is not making it to the No. 1 GP. This is measuring the "voltage drop" of the all the connectors combined, along with the wiring and the GP relay. It's the first step to determining if any of them are bad, but it doesn't tell you which one.

In your case, with the GP Relay engaged, what should read on your VM as zero volts (in an ideal world) is showing 2.5v "lost" or "eaten" by various connectors -- I say connectors, because it's most likely a connector and not the relay itself or wiring, though those are possibilities too.

You can repeat your voltage drop test, moving the negative meter lead "upstream" (toward the battery) and re-engaging the GPs. First move the lead to No. 4 GP and repeat the measurement with the GP relay engaged. Still 2.5v or close? Move the negative lead to the GP relay's OUT terminal, repeat. Then the relay's IN terminal. Testing under load at each test point (GPs being powered, that is). You are probably going to find that you have significant voltage measured until you get to either side of the fusible link, or either side of one of the spade terminal inline connectors.

Philip has written about the battery-to-chassis ground terminal being a weak spot too. That isn't something you'd catch with the above test; you'd have to move the leads to (positive meter lead to engine block, negative meter lead to battery terminal's post).

. . . until the GP relay switches again. At that point, the voltage shoots back up to 12v.

When the GP relay opens up (turns off), no voltage is passed on by it to the GP bus. In effect, the GP relay "eats" all the voltage -- prevents it from moving on. Therefore, you should see full battery voltage on the VM, as the GP No. 1 terminal now becomes pretty much ground potential. This is normal.

Also, should I be able to feel any warmth directly around the glow plug socket while/after they have been engaged?

No. The plug's tip gets hot; the plug body doesn't (much), and the cast iron quickly absorbs the heat.

The injector sprays a pattern of atomized fuel over the GP which is glowing. This is supposed to make it easier to light off, but isn't itself enough to do the lighting.

[Old Smokey]

Thanks for the advice Al. I'm still confused about what might be going on. From the above mentioned test, it seems that somewhere along the glow plug circuit, I am losing 2.5v, right?

Aren't the remaining volts enough to power the plugs? Would that small drop be enough to effectively hinder the plugs from glowing?

Another thing I noticed today, while starting (HUGE smoke cloud and multiple stalls by the way): the afterglow controller clicks off immediately after I give the engine a little throttle (trying to maintain an idle). From the other posts on this topic, it seems that the Auto Glow system normally remains on for about 20 seconds after starting the vehicle. Philip mentioned that the AG unit might be wired to read RPM from the alternator? If that's the case, my revving is causing it to shut off, or it's just a faulty unit, or...I don't know.

When starting today, it definitely seemed like it was missing for about 10 seconds, then warmed and smoothed out. Tons of smoke, too.

What are your thoughts on this Al? My first inclination was to put a larger, new battery in but this seems like it would have little effect based on what I know now. The one in there is over 3 years old, and it's a smallish no-name brand. Then again, it reads 12v or better. I don't get it.

I'll keep testing along the GP circuit to figure out where the drop is occurring and I'll pull the plugs and test them individually.

Any more help would be much appreciated.

Andy

[philip]

-snip- Philip mentioned that the AG unit might be wired to read RPM from the alternator?

Nope ... not this "philip". Both TypeI (SD22) and TypeII (SD25) are "timer" and coolant temperature controlled. There is no RPM involved.

(To date ... "we" have yet to see a tachometer)

[Nissan_Ranger]

Thanks for the advice Al. I'm still confused about what might be going on. From the above mentioned test, it seems that somewhere along the glow plug circuit, I am losing 2.5v, right?

Aren't the remaining volts enough to power the plugs? Would that small drop be enough to effectively hinder the plugs from glowing?

Another thing I noticed today, while starting (HUGE smoke cloud and multiple stalls by the way): the afterglow controller clicks off immediately after I give the engine a little throttle (trying to maintain an idle). From the other posts on this topic, it seems that the Auto Glow system normally remains on for about 20 seconds after starting the vehicle. Philip mentioned that the AG unit might be wired to read RPM from the alternator? If that's the case, my revving is causing it to shut off, or it's just a faulty unit, or...I don't know.

When starting today, it definitely seemed like it was missing for about 10 seconds, then warmed and smoothed out. Tons of smoke, too.

What are your thoughts on this Al? My first inclination was to put a larger, new battery in but this seems like it would have little effect based on what I know now. The one in there is over 3 years old, and it's a smallish no-name brand. Then again, it reads 12v or better. I don't get it.

I'll keep testing along the GP circuit to figure out where the drop is occurring and I'll pull the plugs and test them individually.

Any more help would be much appreciated.

Andy

I eliminated the timer circuit completely and have no problems whatsoever with the manual pushbutton that took its place. I also doubled up the wiring size of the relay feeder. In the warm weather, I glow the engine for about 15 seconds and post glow if (and only if) there is any roughness until the idle is smooth. In cold weather, the times pretty much double and post start glow is always required...

You mention that your battery reads 12v or better. 12v is pretty much a dead battery reading. 12.6 volts is a fully charged battery. You need to be fairly precise when measuring and evaluating voltages, both static and drops. I put the biggest battery in terms of CCA and capacity that will fit into the tray when I replace a battery for my SD22; scrimping on battery size and quality never pays off with a glow plug assisted diesel engine.

Problems with the SD22 glow plugs apart from the timer are mainly connection problems or open circuited plugs. Personally, I have had a main power relay fail to conduct full power when energized as evidenced by a huge voltage drop through the relay. I have had individual glow plugs with poor connections at the terminal nuts which are made of aluminum on two of my engines. A clamp style ammeter will show the amperage draw of each plug. If such a meter isn't available and you are capable, a common automotive style ammeter rated to 60 amps will read the amperage if you connect it inline with each individual plug and it's feeder. By the way, poor connections (Not totally failed ones) will warm up connection points significantly while electrical current of the amounts used by the SD22 glowplugs is flowing through them. If you have the same kind of crappy aluminum nuts as I had on my glow plugs, I would replace them with brass nuts. Be sure to clean the threads on the glow plugs well as aluminum oxides are tough to remove and don't conduct electricity worth a poop. I had one that tested open until I thoroughly cleaned both the connector threads and block threads of the plug.

For fun and testing, you could try totally bypassing your OEM glow plug feed system with a manually operated Ford style fender mounted starter relay hooked up to feed each glowplug with it's own 8 gauge wire. Trip the Ford relay for about 15 - 20 seconds, then start the engine before releasing the ford relay at the end of the glowtime .

Hope this is of use...

Regards,

Andy

[asavage]

I am losing 2.5v, right?

Yes. My "ouch" point is anything over 0.5v .

Aren't the remaining volts enough to power the plugs?

You can start it, right? But not very well. That pretty much answers your Q

Would that small drop be enough to effectively hinder the plugs from glowing?

They'll glow at lower voltage, but not nearly as well.

Another thing I noticed today, while starting (HUGE smoke cloud and multiple stalls by the way) . . .

You might have a hole in one of your GPs. There's a couple pics of holed GPs on this site.

Philip mentioned that the AG unit might be wired to read RPM from the alternator?

Someone else (redmondjp?) said that they thought it must, based on what they were hearing, but that isn't reflected in the wiring.

I'll keep testing along the GP circuit to figure out where the drop is occurring and I'll pull the plugs and test them individually.

Good ideas.

I agree with pretty much all of what nissan-ranger said, except I don't advocate adding a Ford-style solenoid and bypass wiring: diagnose the problem first, do mods later. Connectors, fusible link, GP bus terminals, GP relay, battery ground to chassis & engine. If those are in good shape, and the GPs are not blown, you have enough heat.

[Old Smokey]

Apologies to Philip about the misquote. I reread what you wrote:

The AutoAfterGlow Timer does register a "running" signal from the alternator. From this signal, the glow plugs on the SD22 are kept ON for the first 30 seconds of engine running. RPM is not measured rather, alternator output is registered.

What would be the cause of the AG Timer shutting off immediately after starting?

Ok, moving on. Anyone care to speculate as to whether driving with no thermostat installed would lead to hard starting?

Previous owner proudly explained how cool running the engine was. Indeed, the few hundred miles I've put on it since purchasing have never gotten the temp gauge up more than 1/5 of the way. Lo and behold, the thermostat housing was unexpectedly vacant! I can only imagine what kind of stress the engine was under, never reaching operating temperature.

Time to pull the plugs on this one.

Andy

[Old Smokey]

Well folks, it's gone from bad to worse.

This morning, I had a truck that would run. It was sick, but it ran.

Now...I might be on foot for a while.

I was able to remove glow plugs 1, 2, and 4 with little drama. Maybe I'm just dim, but I can't figure out how to get any kind of tool to #3. That darned oil cooler hose is hogging all the work room! If there's a better way to get those circular glow plug nuts off other than using needle nose pliers, feel free to chime in. So, I thought I would remove one end of the oil cooler hose to get access to #3. This necessitated draining the coolant.

The drain cock on the radiator was missing its flange, so I blindly took some vise grips to it. Whoops. Damaged the soft brass housing. Drain cock won't budge.

On to the engine block drain: a stuck bolt. Sprayed some penetrant, huffed and puffed to no avail. That thing is on there and access to it is terrible. Judging by the coolant the color of the Mississippi, and the bad drains, I'd say someone left it in there awhile.

So now I'm stuck. My next thought was to remove the radiator (might have to anyway after damaging it), try to pry off the oil cooler hose when it finishes dripping and get at #3. This sucks.

On a related note, the glow plugs that came out were caked with soot and carbon. One looked a little pitted or worn down. Could it be that because there was no thermostat, the engine never got up to operating temperature, thereby not being able to burn off these fuel deposits?

Onward ho. Which one is the deflated emoticon?

Andy

[redmondjp]

Probably the easiest way to get the coolant out is to disconnect the lower hose from the radiator--the coolant will gush out like no tomorrow since you don't have a thermostat installed.

I have removed all of the nuts from the glow plugs when I had my truck, without disturbing anything else--yes, you have to be creative and yes, you have to use pliers (yes, it leaves marks on the knurled nuts )

Now that I think about it, I pulled all of the glow plugs out and inspected them as well as tested them to verify that they actually did glow (as shown in one of the 'glow plug' or 'hard starting' topic threads here somewhere).

And I didn't have to take any of the cooling system apart to do it. I think I used a 12mm box-end wrench on the glow plugs IIRC.

[Nissan_Ranger]

Sorry to read that your problems just got worse...
No thermostat in one of these engines is bad news. My first SD was subjected to that in the Canadian climate for 7 years by the original owner. Ruined the rings. All the rings were pinched into the grooves by black, grungy carbon and the cylinder walls were glazed. Oil consumption was high as was blowby. And it was hard to start just like yours. My first clue was when there was no heat from the heater in the first cold weather I drove in. You need to clean things up, get a thermostat into that baby, change the oil and filter more frequently for a few changes, and run some good quality fuel conditioner through the engine. With luck and if you're in a warmer climate than I am, your engine might not have suffered as much damage as mine had.

Al... Looked my earlier post over and I think your prolly right to not to go with sticking in a bypass system for testing. I've been working on diesels for over thirty years now, mostly farm tractors, and doing stuff like that for diagnostics comes to me like a duck takes to water, especially when I have a gut feeling on a diagnosis. That doesn't make it the best thing for someone else to try unless they have my kind of 'time in' on that sort of work... And if they DID have that 'time in', they wouldn't need my help anyway:-)

Regards,

Andy

[Old Smokey]

Thanks to everyone for all the advice. I've calmed down a little.

I was able to get the #3 plug out by removing the altitude compensator and using a little needle nose precision. I wish I had my camera working but it just went kaput so I'll try describing it.

All four glow plugs were seriously fouled. I mean COVERED with a sooty deposit. A couple were pitted with little pock marks. I haven't gotten around to doing an individual glow test yet. Don't know if it's worth it. They look pretty bad and plug #1 failed a continuity test, while the others passed.

Al, could this be where the 2-3 volt drop was occurring?

In other news, the radiator is out and fortunately not damaged, just filled with what looks like a mud puddle.

I still don't have a clue how to get out the engine block drain plug. Just can't get the leverage. Want to try the propane torch but I'm not sure about all that petroleum near by...

I have a theory about the order that everything happened. See if it makes sense to you:

Previous Owner, Johnny Q. Diesel, finds that he can't get the coolant drained due to a: 1)broken radiator drain cock and 2)stuck engine block drain bolt.

"No big deal," he thinks. "The engine still runs fine." Time goes by and the accumulated material in his coolant causes the engine to run hotter. He adds more water, effectively introducing more minerals into the cooling system. Eventually it runs so hot that he has a head gasket leak. After a new gasket (previous owner mentioned he had gotten one in the past 6 months), he removes the thermostat and the engine has run cool since.

Unfortunately, the cool engine was never getting hot enough to fully consume the fuel it was given. Over time, this unburned material caked up on the glow plugs so much that they weren't providing adequate heat at cold start. Hence my original problem of hard cold starts. Now I now I was only glowing on 3 cylinders, and probably poorly.

What do you think?

More importantly, what now??

2 questions:

How do I get the stuck block plug out?
AND
Is it worth testing the glow plugs or should I scrap them all?

Thanks again everyone.

Andy

[asavage]

Al, could this be where the 2-3 volt drop was occurring?

No. The GPs are wired in parallel; they could all be bad, and it wouldn't be the cause for your 2.5v drop. In fact, with all four GPs pulling current, the 2.5v drop will get much worse: more load + poor connection = more voltage lost across that/those bad connectors.

the radiator is out and fortunately not damaged, just filled with what looks like a mud puddle.

Disconnect one or both of the heater core hoses, run a garden hose to the heater core. It tends to hold a lot of sediment.

I still don't have a clue how to get out the engine block drain plug. Just can't get the leverage. Want to try the propane torch but I'm not sure about all that petroleum near by...

Can it even be done with the starter installed? Until you mentioned it, I had no idea where it is, but with help from the FSM, it looks like it is obscured by the starter:

"No big deal," he thinks. "The engine still runs fine." Time goes by and the accumulated material in his coolant causes the engine to run hotter.

Hmmm. These engines don't really run hot; they can reject heat pretty well (just ask Philip and Galen, who do not run a fan much even in deserts). The green coolants use a silicate coating that deteriorates in two-four years and around 30k miles in gassers. After that, no corrosion inhibition, electrolysis occurs, oxides form, but it'll still cool pretty reasonably.

Unfortunately, the cool engine was never getting hot enough to fully consume the fuel it was given. Over time, this unburned material caked up on the glow plugs so much that they weren't providing adequate heat at cold start.

Well, the carbon on GPs is more a function of fuel quality, injection timing, and injector spray pattern than heat. Also, when you test them, you'll find that they really do glow, and any fuel-like substance is not going to stay on them long. Too bad you don't have a camera, but if you have a coating that is more than, say, .030" thick, I'd be looking at the fuel quality and the injectors, if the problem persists after sorting out your poor supply current issue(s).

Now I now I was only glowing on 3 cylinders, and probably poorly.

One or two bad GPs is the first or second thing we look for in SDs that fire up rough and with white smoke.

Is it worth testing the glow plugs or should I scrap them all?

That's a good Q, one I cannot right now answer.

For years, if a GP glowed, I used it. Until the LD came along. Now I'm not so sure that the heat output is consistent across the GP's life-cycle. So I am going to fence-ride on this one: maybe just change the bad one for now, but keep in mind that you might want to replace them all at some point.

By and large, failed GPs really aren't all that common, outside of bad GP controllers keeping them turned on too long (very common on the GM 5.7l, 6.2l, and early Ford/IH 6.9l engines). But when people go to manual control of the GPs they tend to get over-glowed; people advancing the injection timing tend to overheat the GPs even when they're not on; and engines in poor condition (low compression) have to run more glow cycles or have repeated attempts at starting, which increases the amount of total time per year the GPs run. All of these contribute to eventual GP failure.

I'll also leave the block drain plug removal advice to someone who has actually removed it on an SD; I've had my LD ones out, but access appears to be worlds better:

[Old Smokey]

Thanks for your detailed reply Al.

I tested the glow plugs individually. As mentioned above, three of four had continuity. They all took quite a while to glow, and only the tips got red. I went ahead and ordered a set from Napa; should be in tomorrow.

Following the FSM, I tested the water temp sensor and glow plug relay. Both checked out. Have yet to check the afterglow timer, which I was having problems with - it seemed to shut off immediately after starting. Will do that as soon as I get the thing hooked up again.

I guess at this point I have 2 questions:

1) I would like to replace the glow plug harness wire with something newer, less corroded and more substantial. I would also like to run a parallel wire from the the hot relay terminal to the glow plug bus. These seem like fairly straightforward modifications that could reduce the "eaten" voltage I was experiencing earlier (in the neighborhood of 2.5 volts lost). Anyone have any advice, tips or suggestions for going about this wiring?

2) Also, Al mentioned above that:

the carbon on GPs is more a function of fuel quality, injection timing, and injector spray pattern than heat.

Which of these, in your experience, would be the most likely candidate? If it matters, the timing marks are aligned and the smoke screw is still factory wired. The primary fuel filter has a nice layer of blackish sediment at the bottom of it. I have a replacement that I'll install once everything else is in place. Hmm... Suggestions?

Thanks,

Andy

[Nissan_Ranger]

The more details you reveal on this engine, the more it looks like an abused pet that's never had any TLC until you got hold of it. The neglected cooling system, fouled glow plugs, and fuel filter all point to a complete failure to maintain the engine. I've run into this kinda neglect in the field and in the end, the engine in question has done the best when I replaced all fuel filters, oil and oil filters, serviced the sediment trap, pulled the injectors for testing/tune-up, set valves and checked timing. Cooling system flushed (Use an air injecting flushing tool for the heater core; hosing alone won't get out the sediment) I would take out the starter for access to the block drain; that would give me the excuse to send it off to the autoelectric shop for tuning up. Make sure you use a 6 point heavy duty 1/2 inch drive socket on that block plug; a regular 12 might slip and strip off the corners. Forget the propane torch; it won't put wind under that problem. In any case, that's a tapered pipe plug and will come out without heat when you can get proper purchase on it.

Regards,

Andy

[Old Smokey]

Ranger:

I agree. You certainly can't judge an engine by how it sounds. This thing is one sick puppy.

I was able to get the block plug out, drain the muddy water and get to work on the glow plug system, which I'm in the middle of now. I want to parallel wire from the battery + post to the GP relay, then from the relay to the new, thicker gauge glow plug bus. As mentioned earlier, I was experiencing lost voltage through the GP circuit and other posts have mentioned this mod to bypass the spade connectors and fusible link.

If anyone has any advice on running these parallel wires, please let me know. I'm supposed to be moving out of my apartment right now, so I've gotta get after it.


Andy