Page 1 of 1

Some Nissan diesels contrasted

Posted: Fri Feb 28, 2014 9:49 am
by asavage
While not off-topic, I couldn't in good conscience put this in one of the other forums.

Cal recently contacted me asking about Nissan diesels. When my email response got to over an hour of typing, I figured I may as well put it here instead. Here's what I wrote:
Cal wrote:Noticed url in note regarding Nissan. Knew folks with small diesel powered Nissan pickups that raved about the powertrain. These being 80's. Appears to be excellent candidate for powertrain conversions. Do you know how available engine/transmission parts are for such? How rebuildable are these? Some Japanese overhead cam engines tended to wear out the heads where camshafts were located?
Nissan's been making diesel engines for a long time. They have had in production an astounding number of designs, and of the ones that were available in significant numbers in the US market, they are all very different from each other.

CD17: 1983-86 Sentra had an optional CD17 4-cylinder OHC traverse diesel with twin timing belts. In other markets, I've heard this engine came as a CD20 and sometimes with a turbo. The CD17 is a very thrifty engine but is a small-displacement engine and only suitable for light vehicles. As with other belt-drive valvetrains, people do neglect to replace the cam belt and the camshaft & head can be damaged if the belt breaks. The head is aluminum.

LD28: 1981-83 810/Maxima had an optional LD28 6-cylinder Inline diesel, rear-wheel drive, non-turbo in all markets. This engine can still be found in Albin cruisers (boats), agricultural pumps, gensets, etc. It's OHC, 2.8l, and a very smooth engine. In other markets (Australia, NZ, etc.) this was offered in a 4-cylinder variant LD20 with a different bore and sometimes a factory turbo. The LD28 shares critical block dimensions with the gasoline-powered L24/26/28 in the 240Z etc. cars, and someone has successfully converted an LD28 block into a gasoline engine in the past. The LD engines have a chain-drive OHC and there is never a problem with their heads or cams. The head is cast iron.

SD16/22/23/25/33: 1981-86.5 (US: model 720 pickup) or 1987 (Canada: model D21 pickup) had an optional 4-cylinder diesel, the SD22 (and SD33 Six-cylinder in IH Scouts/Travellers, and many marine applications). It's a pushrod engine with agricultural roots, and its agrarian nature does tend to show. It sold in some numbers '81-82 in the US as the SD22, in lesser numbers in 1983-4 as the SD23, and rarely as an SD25 after that. In agricultural use there was a 3-cylinder variant that I've only seen once, the SD16. This engine was fairly popular in small boats going back to the early '70s, and you still see them in forklifts everywhere. They are not as refined of an engine as the other two above, but do seem long-lived for all that. They are the most plentiful of Nissans US-diesel offerings at this time. The head is cast iron, and cracking of the head is a common problem, though they can be repaired by a competent head rebuilder.

I've owned examples of each of the above vehicles (I had the Maxima for 14 years). They're all good engines in their own way. I tend to like the LD28 best because it's a Six and it has great power for a non-turbo diesel of that vintage.


First, know that the wiring connectors on the glow plug systems on all Nissan cars are trouble spots. I've repaired countless no-start Nissan diesels by merely repairing GP system connectors. Nissan used 8mm spade terminals on 10 gauge wires commonly carrying over 50A loads, and spade terminals just do not last in that situation for three decades.

Moving on . . .

As the SD engines were used in so many different applications (forklifts, boats, ag pumps, and vehicles), rebuild parts are by-and-large readily available for engines in non-automotive trim. I phrased that carefully, because some automotive-specific accessories are beginning to become difficult to find, notably the leather pneumatic governor diaphragm in the Inline Kiki-Diesel injection pump on the SD22. The engine can be converted to an Distributor style IP and there appear to be no parts availability problems with that IP, but the Inline IP is more durable. That's just one example.

Piston & liner kits for the SDs are readily available and fairly inexpensive. You do not typically bore and SD, as the liners are a light press fit and inexpensive to have replaced. The SD is a dry-liner engine.

The SD engines wear fairly fast if not maintained, and many/most I've seen are oil burners. They will run when severely worn, though of course they smoke badly when in that condition, and they can be difficult to start when worn, as are all worn diesels. Most of the SDs I've encountered have needed rebuilding. When unworn and in proper tune, the SD is a very clean-running diesel, exhibiting little smoke.

The US market SDs with the Inline IP all have a fairly elaborate (for the day) control system, with an electric motor that drives a quadrant gearbox to put the IP into one of three modes: Stop-Start-Run. The control box has a soldered relay that fails and can be upgraded, but most people -- lacking the knowledge or ability to solder in a relay -- convert to a "choke cable" system that is much inferior but also much simpler and easy to make-do with. To each their own. The electric motor never goes bad, though they do with age get noisy and squeal when you turn them on or off. The system shuts down the engine if the oil pressure gets too low, a protection system.

The FS5W71 transmission behind most 720/D21 pickups has a diesel-specific "bellhousing" (really, the front case half). It can be transferred onto a gasoline version of the trans as a bolt-on. This can be handy, as the trans when used behind a SD can eat the front mainshaft and countershaft bearings -- it's a very common problem. You want to change the oil in those gearboxes frequently and use premium synthetic oil. A lot of 720 pickups were scrapped due to trans failure.

Reverse gears are fragile, so you don't want to abuse reverse.

Parts for the trans' are readily available, and bearing upgrades are possible too.

The SD was never offered in the Nissan pickups with an automatic. However, the six-cylinder version, the SD33, was offered in IH Scout/Travellers up to 1980 and they had an optional Chrysler Torqueflite AT behind them -- a special version, and rarely found in the wild.

The LD28 was not offered in many different applications as the SD, and parts are harder to find. The LD is not a linered engine, so you do bore them to rebuild. Gaskets are getting hard to find in the US, but still available overseas. The head gasket has been reproduced in copper. The IP is readily rebuildable by any competent injection pump rebuilder.

The LD engines I've encountered have never needed rebuilding. Given clean oil, they mechanically run "forever". The LD's weak spots are the IP belt and the IP itself.

The Injection Pump is a Bosch-licensed copy of the venerable Distributor style IP in use by millions of diesels worldwide. It's a six-cylinder version, of course. It has an internal lift pump that wears out rapidly, causing hard-starting, fuel starvation, and most importantly a lack of timing advance. As a result, the engine runs with retarded timing at mid & high RPM and smokes. LD28s in Maximas tend to be very smoky engines, because they pretty much all need to have their IPs rebuilt.

The other LD28 weakness is that the IP is driven by a toothed belt with a replacement interval of around 80k miles. It's hidden under a steel cover at the front of the engine and many mechanics don't know it's even there. When the belt strips (it has very load loading) the IP goes out of time, the engine won't start, and the car goes to the scrapper. ALL of the no-start LD engines I found have had either GP connector issues or a stripped IP belt. ALL of them.

The LD28 was sold with a manual transmission, the same FS5W71b mentioned above for the SD engines, and that trans has the same weaknesses as mentioned above. The LD, as sold in the 810/Maxima, was offered in the US with THREE DIFFERENT versions of an automatic transmission:

1981: 3N71B = conventional 3-spd AT
1982: L3N71B = 3-spd w/lockup converter
1983: L4N71B = 3-spd w/lockup converter + OD

Although opinions vary, I have yet to encounter a failed version of this AT. They seem to be durable even when abused horribly. The last year version, with the overdrive, is the least reliable if the fluid isn't kept clean, but also gives the best mileage.

The CD17 did not sell in large numbers in the Sentra in the US, and is the least-encountered Nissan diesel here for that reason. The seem reasonably durable and quite thrifty in the light Sentra. They were sold with both a 4-spd & 5-spd transaxle, but not an automatic. Due to its FWD layout and the lack of a RWD transmission here that will mate to it, I have not heard of anyone using this one for a conversion. A member here said he did mate a CD20T into a Infiniti, a G20 I think.

Cal, I hope this isn't too confusing or detailed -- I just couldn't stop. The summary is that the three Nissan diesel engines that are readily available here have very different profiles and you'll want to evaluate your intended use before choosing one.

There are a number of importers of engines from overseas that will sell you all kinds of other Nissan diesels, and once you get away from the three series above (LD, SD, CD), I don't know anything about them, and can't fairly advise you.

Regards, Duvall, Wash.
Al S.

Nissan Diesel Engines

Posted: Sun Apr 13, 2014 3:29 pm
by Duaneclark
Thanks very much for this detailed discussion. Would like to do more research on the "LD28 IP internal lift pump" concerning repair and maintainence.

Posted: Wed Apr 16, 2014 8:03 pm
by Carimbo
IP internal fuel supply pump (sometimes referred to as fuel transfer pump) is a vane-type pump similar to the vacuum pump on the rear of the LD28 alternator.

Several ways to destroy it:

1. Run w/ clogged fuel filter will cause the vane pump to struggle and cavitate. Pits & heat-scores the vanes, rotor, and stator. Has been known to destroy the rest of the IP by sending shavings downstream. Easily avoidable.

2. Run WVO w/ suspended water also will cause cavitation damage. Can also gum-up the vanes and make them "lazy" to fling outwards = no sealing, especially at low RPMs, like starting, for example.

3. ULSD in an IP designed for LSD. Lack of lubricity can prematurely wear the vane and stator surfaces.

Here is a good visual, internal fuel pump section starts 5-6 photos in: ... +stripdown

It IS possible to rebuilt this area, but it is pretty far into the IP (have to pull almost everything else out of the case to get to it) and would require IP/Injector rebuilding best practices: Ultra-clean work area, copious disassembly/reassembly notes, matchmarks, a few special tools, etc. Helpful to practice on a junk IP first.

Good information on the IP

Posted: Fri Apr 18, 2014 5:47 pm
by Duaneclark
This is good information to know. The pictures of the IP re-assembly, (in the link provided), are very good quality to visualize the procedure. I can't help but wonder, if that if the IP functions well, but the incorporated lift pump is weak, would not an external fuel pump adding pressure to the IP inlet, take care of that problem?

Posted: Fri Apr 18, 2014 6:03 pm
by asavage
No, no, NO!

The function of the VE-style IP's lift pump is not merely to suck fuel from the tank to the IP (which is what the Inline-style IP's lift pump does). The VE-style IP's lift pump is integral to proper injection advance occurring, and its output must be able to crest 105 PSI.

Please read this post for more details. And if you care to, follow the link in that post to read the story of Tom's Toyota diesel that would only run with a pusher fuel pump plumbed in, but with no power. The pics are good too.

continuing onward

Posted: Sat Apr 19, 2014 3:02 pm
by Duaneclark
Many thanks again to our resident genius, Al. Of course only one "NO" is necessary. It looks like the speed of the lift pump determines the pressure, as-in, more speed, more pressure, and then this regulates the timing device, so that a flat fuel pressure coming into the IP would not work correctly. More good information gleaned from the link, is a way to test the lift pump with a vacuum gauge, so those curious as to the condition could get a idea of what they have. Painting a broad brush over all of automotive stuff, such as our desires to operate 30 year old vehicles way past their designed lifespan, I think the original manufacturer's of these vehicles would simply be amazed that there are so many of these vehicles still on the road and being loved and cared for.

Posted: Sat Apr 19, 2014 4:56 pm
by asavage
Sorry for the duplicate, emphatic "nos". It seems like I respond to that suggestion at least once a year. I don't know why I think that yelling No! will slow down that trend, but it felt good ;)

I don't know if there's a vacuum test for the VE-style lift pumps. What I'd really like is a pressure test for the IP. There should be an adapter fitting for the IP housing to which a pressure gauge could be attached, and run up the revs and see if there's adequate output.

There's a bleed orifice on the return line. On the EGR-equipped LD28s, that orifice is further restricted by a solenoid during EGR operation, to advance the IP timing when EGR is active. Of course, with little lift pump output, this timing boost does not occur.

I've said it before, but again: I think most VE-style IPs have out-of-spec (worn out) lift pumps. From the several I've worked on, they "all" seem to be worn out. That's why the black smoke on otherwise perfect engines -- I'm talking LDs here, not SDs.

An observation about black smoke

Posted: Sat Apr 19, 2014 6:30 pm
by Duaneclark
I've noticed on a couple of my Maxima's, when they were at low altitude, I couldn't see any black smoke, now at 6500 ft., they only do the black smoke when you get on the accelerator going up a hill. The same for some Isuzu diesels. I think that's normal for these older examples. Naturally at elevation there's not enough air. A condition of no smoke at all, even at low altitude would indicate a healthy Injection Pump and timing, I'm guessing. Regarding testing the pressure the lift pump generates, you would imagine that at an official factory service facility, that's a procedure that they would perform on a test bench, if it can be done, some one in that capacity should know if that test is performed.

Posted: Sun Apr 20, 2014 9:46 pm
by Carimbo
asavage wrote:The function of the VE-style IP's lift pump is not merely to suck fuel from the tank to the IP (which is what the Inline-style IP's lift pump does). The VE-style IP's lift pump is integral to proper injection advance occurring, and its output must be able to crest 105 PSI.
Also remember that the internal advance mechanism works off of the pressure differential between pre- and post-internal vane pump. So if you are feeding the IP (and therefore internal fuel pump) pressurized fuel, you may even be reducing that pressure differential and hence reducing further an already weak dynamic advance curve.

Look at this LD28 FSM exploded dwg of the IP; specifically the "Regulating valve" which I could only describe as a pressure-bleed valve (adjustable by a friction-fit post that sets the preload) inside a screw-in housing. Some VW folks have tinkered with this and have been able to rejuvinate (read: "compensate for worn-out) the vane pump pressure output by gingerly tapping the inner post downward, thus boosting the housing pressure/pressure differential that the internal advance mechanism sees. (They have also been able to tap it in too far, requiring removal to tap it back from the other side and start all over again.) See Topic: Internal injection pump pressure ... 36&start=0
asavage wrote:What I'd really like is a pressure test for the IP. There should be an adapter fitting for the IP housing to which a pressure gauge could be attached, and run up the revs and see if there's adequate output.
RE: Pressure measurement: Here also in the VW IDI world (strength in numbers!) they have come up with a method to modify an IP fuel "OUT" banjo (see prev. exploded dwg. near the top, between "Maximum speed adjusting screw" and "Control lever." The unlabled banjo bolt and orifice) fitting to connect a pressure gauge. See prev. mentioned Topic: Internal injection pump pressure ... 36&start=0 for helpful photos.

Here it starts to get a little tricky. What are the internal pressure specs for the LD28 Zexel licensed IP? (Likely different from the VW or even Toyota versions) How does the IP-mounted EGR/IP advance controller rheostat figure into this? (Controls a downstream IP fuel return restrictor solenoid)(This is what Al refers to above)
asavage wrote:There's a bleed orifice on the return line. On the EGR-equipped LD28s, that orifice is further restricted by a solenoid during EGR operation, to advance the IP timing when EGR is active. Of course, with little lift pump output, this timing boost does not occur.