Looking forward to it!

 

No, I didn't replace the blue O-ring. With the holiday there was no way to source it and land it that fast, and I figured your comment about that ring never seeing high pressure fuel was accurate. It would be the last man standing and as bad as the condition of that ring is, it would be known for leaks/fire/whatever if high pressure was what it was containing.

Could be wrong but I knew it was safe to operate before so I ran with it. I would still absolutely recommend replacing it though, and since I have other HPFPs I will do so. In fact, will probably do so on this one after it has been proven, since I'll want to test my others when the time comes.

Regarding torque, I didn't ID a spec but I did fiddle with an HPFP from a junker, whose bolts had not been touched. Net, they are easy to unscrew with a T-handle screwdriver -- there's an initial pop, but they're not torqued to 100 ft-lbs or anything like that. A "strong two-fingers-tight" using the T-handle screwdriver seems to be where to go. I would like to quantify that if someone can give me an idea how to, since I do have torque wrenches.

 

I appreciate your continued feedback on this thread, and look forward to your report on cold start from this morning.

As for torque value on the solenoid screws... Since the oring is providing the seal, the screws only need to be snug enough to not back out. There only needs to be enough pressure to seat the oring.

 

That's good feedback on the solenoid screw torque. Thank you. Makes sense -- very much appreciated.

 

Ok - cold start quick overview.

- The car was on a hill (my driveway), nose up, overnight and when I got outside it was 18 degrees F. That pretty much puts the "cold" in cold start, especially for this geography (Atlanta area).

- Before startup, fuel line pressure was 68 PSI, coolant temp 32F. Hit start, it turned quickly just under 5 times and revved right up. Vroom vroom, no misfires, no stumbling, no stammering. Let it idle for 2-3 minutes from start, within which it dropped to a normal idle RPM on its own. Decided it was stable. Fuel pressure did not jump to normal range -- topped out mid-80s. Time to drive.

- Drove. Drove a lot. Half engine light and found a pending P15DF. Fuel pressure would go above 100 but not much more and the half engine light would not clear. Revving the engine while stopped, even while well into normal operating temperature, didn't get the HPFP to pop into normal range. I became highly suspicious that the DME had put the car in limp mode and called it a day, so after nearly an hour of driving I pulled the car into a parking lot, cut the engine, and restarted it. Boom, almost 680 PSI.

- Pushed the car around a lot. Half engine light stayed along with the pending P15DF. But had my performance from last night back. Went and got some food, let the car sit and cool off a bit.

- When I returned coolant temp was down to 167 and startup yielded a pending P0087 to park alongside the pending P15DF, which was staying pending (meaning not being confirmed). Fuel rail prssure was back down to the 80s. Gunning the engine while parked didn't make the HPFP jump to normal range.

- Drove for a bit. Decided to pull into a lot, shut it down, crank it up and gun it right after startup. Wound up with a pending P15DF and P0087 but giving the immediate gas got PSI above 1000, and when it dropped back it dropped to normal range (high 600 - 700). Stayed that way. A later check revealed that the P0087 had cleared up but the P15DF was hanging around, still pending. HPFP was still being occasionally flaky after a restart, but for the most part, once it engaged it stayed engaged.

A later clear of the codes vanished the P15DF, but the P0087 returned. The half engine light did not return. Once the HPFP got engaged, it stayed engaged. Drove around for another hour, pushing the car in different ways, and things held. By "held" I mean "felt like a Mini, the only thing left to do was put wings on the car."

I have a bunch of screen shots but I don't think they tell much more. The upshot seems to be that the HPFP needs to "warm up," whatever "warm up" means, and once it's "warmed up" it stays ready to go. I am highly suspicious that the DME gives up on it before it is warmed up, and that the presence of pending fault codes may predispose it to keep things in limp mode. Gunning it quickly after starting, before it has long enough to throw the half engine light, seems to preempt the half engine light *if* the HPFP is ready to go.

The wobbly bits of this are consistent with what fully defined my pre-repair experiences. The interesting part is that what is not wobbly, and not part of my pre-repair experience, is the fact that *now* when this post-rebuild original HPFP is given opportunity to hit its stride, it becomes a no-holds-barred can't-stop-it workhorse.

My read is that this is either a break-in thing, an HPFP-needs-additional-work-not-done-yet thing, or that there is an additional DME and/or sensor variable in play.

Thoughts?

 

Re the P15DF in particular, this link (a PDF) is of interest -- search for the code in the doc, read the whole row. What it says about limp mode may explain some of my experience.

https://www.bimmerfest.com/forums/at...5&d=1425058449

From the doc:

Possible apparent symptoms:- Loss of power
Breakdown notice:The engine reverts to its limp-home program, continued vehicle operation is possible but drivability is restricted, because power is reduced the driver should refrain from passing maneuvers.

Also from the doc:

"The fault is recognized when the pressure level in the high-pressure system is 3.5 bar lower than the specified value."

Note also that the signal source is described as "Rail-pressure sensor voltage signal." So it's a separate measurement from the HPFP itself or something back on the LPFP. The same signal source is used for the P0087 according to the doc.

During the low power portion of this morning's drive I had to back off of merging into highway traffic, sub-speed limit (65MPH), because of the experience it describes. That "power is reduced" is not an exaggeration -- it was a real problem, not just an annoyance.

 

Ok - after round 2 of driving today, I have some observations and a theory.

More focused and on topic: if I can prove that the HPFP is not the issue in my car, then that probably means the HPFP rebuild was spot-on. So on from there...

Observation 1: historically, my Mini may recover mid-drive from a half-engine light unaccompanied by a P-code.

Observation 2, made today: my Mini will *not* recover mid-drive from a pending P15DF (Cold Start Fuel Pressure Too Low (Bank 1)). If that's in the mix the car is in limp mode until the code goes away, whether the MIL is illuminated or not. Currently, I'm stopping the engine and clearing that pending code manually as part of testing/troubleshooting. This is obviously unsustainable as a normal-use approach.

Observation 3: As previously noted, I replaced my LPFP, and the in-tank fuel filter, as part of troubleshooting. Minimum acceptable low pressure, engine not running, is 5 bar/72.5 PSI (Bentley manual, p. 160-10). Just before initial cold start under *both* LPFPs I would routinely see as low as 68. In monitoring both LPFPs (I did the replacement back in either late August or early September) their general behavior, pressurewise, appears to be the same. My best-guess takeaway: either both LPFPs are broken, the sub-minimum readings at cold start are normal despite the spec, or the sub-minimum readings at cold start are being misreported.

Theory, part 1: the refreshed HPFP is capable. I think this is the case because at least post-warmup, it doesn't flinch. (note: still have to figure out what "warmup" means because the only thing allowing us to learn about the HPFP is the fuel pressure sensor, and the sensor is suspect.)

Theory, part 2: the P15DF is the result of genuinely low cold start pressure *reporting* by the fuel pressure sensor, and has nothing to do with the HPFP.

Theory, part 3: having the P15DF "on file" means the car is going to stay in limp mode, basically shutting off the HPFP and masking its ability to keep up. Same may apply to having a P0087 "on file" (Fuel Rail/System Pressure - Too Low), but I would have to do more methodical testing than I have been to date to confirm this.

Theory, part 4: the P0087 and the P15DF are both misreports (meaning the fuel pressure is being underreported by the fuel rail pressure sensor).

If I am right about this, the next step is to replace the fuel rail pressure sensor, meaning replace the fuel rail, as suggested by MiniToBe.

Can anyone else out there who is wintering right now check their fuel pressure reporting in the AM, after the car is turned on for the first time (BUT the engine not yet started)? That's when my car is reading sub-72.5 PSI fuel pressure. Since this has been consistent across two LPFPs, I want to compare it to not-my-Mini's fuel pressure sensor. Maybe we can "prove" that mine is wonky before I drop $400 on a new fuel rail.

I'm using an OBD adapter (iCar -- Bluetooth LE) with a phone app (OBD Fusion) to do the monitoring.

 

Injectors - I haven't heard of any going bad but always a possibility.

Also it's worth noting that the solenoid can be removed without taking the pump off the car, remove the negative battery cable, remove both fuel lines to depressurise the pump and slowly remove the 2 bolts. If this side of the pump is the issue it saves time removing the whole thing the lining it back up! Just be ready for a cloth to catch the fuel inside!

it might also be easier to dislodge the solenoid with the pump still attached...mine was a right pain in the *** as I didn't want to use a vice incase I caused damage!

EDIT: So, for some reason despite refreshing the page the above posts after the one I quoted did not load.

good news about the pressure holding, I'll be making some calls today about getting the correct o rings now the holidays are over and will test this for you. If the same happens on my car we know it' HPFP build related...if My car is fine, it' likely a deeper issue with yours.

If it is something that needs to get warm try and pull the solenoid out again and check the o rings, possibly one is not seated quite right and when it gets warm the o ring softens a little and moves slightly, falling into place.

 

A few things (guesses).

I find it hard to believe that the injectors can function at all at around 80 PSI. Seems like it could be bad data. This is just a guess being that the PSI only goes as low as 600 PSI during normal driving.


In regards to the HPFP being limited, I remember reading somewhere that the HPFP goes into emergency mode when a fault happens, and it severely limits the pump, or something along those lines.

Time to replace the fuel rail / sensor.

 

I wondered about that -- I just keep scratching my head to outline a failed-injector scenario that manifests as low fuel pressure sensor readings.

This is really interesting. I think I solved this with the T-handle screwdriver. Trying to use a regular screwdriver to remove the bolts on the HPFP was *not* working, so I removed it and put it in the vise as you saw. But once in the vise, with the T-handle screwdriver, there was so little resistance to removal that I was amazed. So the other day I took to the other genuine-BMW-Mini HPFP in my possession, which came off of a junker, and tried to unscrew one of the bolts with the T-handle screwdriver while holding the HPFP steady with my other hand (all of this action going down on the kitchen counter, ). Worked like a charm.

Still, though, pretty interesting. I kind of like the idea of removing the solenoid without removing the HPFP from the engine.

If you fully and safely depressurize the fuel (I figured out how to do this -- when I remove my fuel lines from the HPFP I don't get *any* spray whatsoever) do you think it's possible to pull the solenoid without even removing the fuel lines? Because it's pulling the fuel lines that gives me a headache. If I could pull the solenoid without pulling the fuel lines off that would make my day.


I was thinking about the seating of the O-rings being an explanation for the cold vs warm behavior. Something during reassembly had me wondering if they seated OK, but I decided not to pull the part back out to check it. I'd be really interested in your thoughts on the "feel" of this during reassembly once you get that far, because it strikes me as tough to know (but important to know) whether an O-ring has stayed seated or slipped and rolled somewhere.

Well, I can offer a little additional detail/clarification here. The Bentley manual shows the low-side pressure to test for as 5 bar/72.5 PSI. That spec is additionally much-discussed in public. The implication would seem to be that that low floor is accurate both in terms of manufacturer spec and field observation.

Moving on from there to the HPFP being limited, the BMW PDF I got from a Bimmerfest thread that documents P15DF (among a ton of other things) in great detail describes the loss of power that is consistent with (1) my observation of the realtime scan data and (2) the actual driving experience, where there is barely sufficient power to properly highway merge anywhere above 55MPH (seriously, it was that bad -- I had that happen and had to back off because I couldn't match the acceleration rate to merge safely at 65MPH with nearby vehicles).

In that low-power scenario the readings for fuel pressure were in the mid-80s. Basically, it falls to the minimum pressure spec, aka LPFP pressure, aka "the HPFP is turned off" from the looks of it. The only time I see readings in the 600s are when the HPFP is behaving itself, and in those situations the 600 range -- usually 620-ish -- is the floor (excepting HPFPs that are clearly failing. I've seen those dip into the 500s and somewhat reliably throw a half engine light when they do so).

Worth mentioning that I have never seen a sub-72.5 reading while the car is actually running. I think the lowest I've laid eyes on while the car was moving was 78, and maybe a 74 at idle shortly after startup.

Still, I think what I have going on is that the fuel pressure sensor reads inaccurately low -- not by much, but still -- under certain conditions, and since a shutdown/startup after some driving seems to get the system to pass the test and flip the HPFP into go mode, the implication is that the pressure sensor has a warmup issue.

Which takes us exactly to...

Ayup. Unless, of course, TurbodTurtle is on to something with the seating of the O rings.

Hey... you know, since the O rings came in packs of 25, I could go put new O rings on my junker HPFP, swap that in, and see what it does... Ok I think I may do this.

In the meantime, still interested to see an actual reading taken as follows:

(1) Cold condition, preferably before first start of the day.
(2) Turn car on but do NOT start the engine.
(3) Read fuel line pressure using OBD tool. This will be the initial fuel line pressure from the car doing its usual initial pressurization.

Can we collect some data points? I can't be the only one out here eyeballing their car's electronic data

 

So this is interesting....

changed the o rings to correct size and material....straight away error code....low fuel rail pressure and torque pro reading 81psi....

Very odd as the previous o rings didn't give me low pressure. Only difference is I lubed the o rings. Maybe something isn't connected right 🤔

The 3 o rings are:

BS011
BS011 Back up (the hard piece under the small o ring)
BS012
BS015

And they need to be Viton material I believe

Edit - BS015 doesn't appear to be correct, it's not thick enough. I have changed it back to the previous one and still low pressure.

I think the one that is the issue is the bigger of the 2, the solenoid is very easy to pull out where as before it took a fair bit of effort to wiggle out.

I can't see fuel leaking out of anywhere and the pipes appear secure. It is going to be one of the o rings inside that's the issue. I'm letting it cool down currently and I'll have it off again. I can get it off in 5 mins now so not too bad.

 

This is interesting. Ok. I didn't replace the bottom ring at at all. I also did NOT replace the hard piece under the smallest O-ring. Am I reading that you replaced that hard piece with another O ring? The hard piece had a diagonal slit in it that I and my dad figured was there to provide some blowby/pressure relief functionality -- I wouldn't recommend replacing it with something of another material+design.

I did lubricate the rings, but I may not have handled it the same way as you. What kind of lubrication did you use, and how much of it? I used petroleum jelly, I lubed the rings before actually performing assembly, and I was very careful to make sure (1) that no jelly got into the holes on the solenoid shaft and (2) that there was no excess jelly anywhere. Like you I had no fuel leaks, and I noted that the solenoid took a little work to wiggle out at initial disassembly (at least if one didn't simply pry it with a screwdriver).

I also have an update -- some interesting data that I think points at the O rings not seating quite right at initial insertion, possibly then causing the warmup issue.

I decided to cold start and aggressively push the car this morning (maybe 18 degrees out). So I cold started -- no trouble, excepting the expected complaint codes and being in limp mode (low 80s PSI). I drove until coolant temp was above 122, then stopped/started the engine. It threw a pending P306F ("Fuel Rail Pressure, Minimum Pressure Fallen Below, Injection Cut-Off") -- that was new. But BOOM, I had normal pressure (700+) and associated performance. During that drive, though, doing maybe 45MPH and going up a shallow hill, the car began to stammer and stumble hard, hard enough that I backed off the gas. The car recovered. A glance at the graph showed that at that moment, PSI did a sharp drop to about 545, and then came back up. A little further along I had a dip all the way down into the 460 PSI range, immediately followed by a spike to above 2000 PSI.

At that point I hit my destination and had the car sit for about 2 hours. Coming back out, the first start brought up the engine but it died immediately. Started it again and had stumbling while backing out of the parking space. Had some stumbling on the road for the first 60 seconds or so with the CEL on, reflecting a P15DF that was no longer pending. It was severe enough that I would have pulled over if I had had a shoulder to do so instead of a turn lane. HPFP was clearly in limp mode based on reported PSI (80's) and vehicular behavior. Got to the traffic light, stopped and restarted the car. The CEL stayed on but PSI headed straight for normal range. Drove home. Performance was good but I felt a stumble or two when I tried to push acceleration so I backed off.

My money is saying the O ring seating issue guess was correct. I'm going to do the O rings on the junker HPFP I have, to gather some observational data about seating the rings, and then later today I'm going to pull the solenoid from the original HPFP on the car to do a reseat op.

Wanted to share the update. Good post and detail, TurbodTurtle, I think we're converging on the same stuff.

 

Do you think Viton is the correct material? Perhaps it's too hard until it gets heated and softens up...?

nitrile is definitely no good though, softer and seems to seal better but as soon as it gets warm it gives in.

I've got mine stripped back down, the BS012 when warm is definitely hard to remove than when cold. The difference is I can't drive with only 80psi so i can't really get it hot. I may try and heat the o rings off the solenoid and squish them a little.

 

 

An update. I've been talking this up with someone who has materials background and navigating the living daylights out of the McMaster-Carr O-ring selections. The Viton rings they sell make sense generically but I have two conceptual problems with them.

(1) Their low temperature tolerance is -15 degrees F. I have walked around in -22F temperatures in Chicago, and that's before wind chill. Seems to me that using the Viton rings with that spec is a great way to guarantee the rings are, sooner or later, going to be harder than you want them to be -- which may speak to the warmup issue. Even if it isn't my warmup issue (if the issue is O-ring seating, for example), it still strikes me as a setup to fail if your Mini is in a particularly cold climate.

(2) The Viton rings are harder in general than other options.

With these in mind I've ordered the following rings from the McMaster-Carr site. These are "Chemical-Resistant Low-Temperature Fluorosilicone" rings, their temp range is -80F to 400F (vs -15F to 400F), and they aren't quite as hard as the Viton (Durometer 70A vs 75A for the Viton).

8333T121 (Dash Number 011, the smallest ring)
8333T122 (Dash Number 012, the middle ring)
8333T125 (Dash Number 015, the bottom ring)

I'm going to replace all three rings, or at least the two we've been talking about, on my junker HPFP. Once that's done I'll remove the original/Viton-rebuilt HPFP, install the junker HPFP, and see how things go. I'll also, once the original HPFP is on the bench, remove the solenoid to see if there's evidence of an o-ring seating issue from install.

I did notice your comment about the larger ring not being a fit so I'll play that by ear. I took digital calipers to a bunch of stuff today from the junker HPFP solenoid and the 011/012/015 inner diameter sizings appear to be correct, as well as either a 1/16 or a 3/64 ring wall diameter (1/16 seems to be spot on).

Since I had the calipers out I tried metric measurements for the heck of it. They don't line up with product I can buy from McMaster-Carr. McMaster-Carr seems to have a ton of stuff so that signals to me that neither the actual rings nor the solenoid are metric-sized.

I also verified that the backup ring at the top of the shaft of the junker HPFP does have a diagonal slit by design, so now I've seen this on two HPFPs. But the slit is crazy hard to see with the naked eye.

Here's a site and a link about o-ring installation. Note that it specifies the use of lubrication to install o-rings, so that's a solid call.

Oh -- the fluorosilicone rings are blue, just like the bottom ring. I'm theorizing at this point that in the as-manufactured HPFP, the black rings are Viton or some custom material, and the blue ring is fluorosilicone. I smell industry standardization in play.

More to come when the rings arrive. In the meantime I'll close with the following observations:

(1) I now have stored P0087 (low fuel rail pressure) and P15DF (low cold start fuel pressure) codes and I'm not resetting them. Regardless, once the vehicle is warmed up to full normal operating temp (200-degree range), I can restart the car and the Viton-ring refreshed original HPFP goes straight to normal behavior and performs admirably.

(2) There is probably a lower temp threshold to get this HPFP to "behave" but I don't know what it is. Since it shouldn't matter, I'm focusing on the change in o-ring materials.

(3) Regarding not being able to get the car hot at 80PSI -- you can get the car hot at 80PSI. I did. Doesn't take much, just drive it and watch your temp with Torque or whatever app you're using

(4) Check out this chart of O-ring material applications/recommendations. This is on the McMaster-Carr website.

 

I made a breakthrough on this today.

(1) Start car. Do NOT give it gas trying to help it. Deal with driving at 80PSI unhappiness.
(2) Drive until coolant temp is 153F (give or take — 153 is where mine was for this). It took maybe 10 minutes of drive around after initial cold start to get to this.
(3) Stop engine.
(4) Start engine. Do NOT give it gas trying to help it.

After that second start the HPFP immediately went to the 600 PSI range, all by itself and without even having to put the car in gear. EDIT: How reliable this is I'm not sure, but it was telling.

I am still definitely seeing signs that the O-rings are at issue, whether a temp thing or a seating issue, but as noted I intend to dig into that when the alternative-material rings arrive.

Hope this helps.

 

https://www.ebay.com/itm/2010-2012-M...haOyev&vxp=mtr

Also check car-part.com

 

Possibly critical update — was just talking to a Mini mechanic — that bell housing-shaped section of the HPFP contains a diaphragm of some kind and is involved in temperature compensation. It may be part of the puzzle I’ve been working, though I have no idea how to approach tinkering with it. Would love suggestions. The solenoid-rebuilt HPFP I have is the original, so all its internals save the two o-rings I replaced are 6+ years and 144K+ miles old.

 

A couple of pics and some details. Just got the "experimental" O-rings delivered.




Above:

Top row, L-R: original top, center, and bottom ring taken from the solenoid on my junker HPFP.

Center row: new "Chemical-Resistant Viton Fluoroelastomer O-Ring, Super-Resilient" from McMaster-Carr. Top and center rings only (did not order a bottom ring of this material). These rings are what I installed in my original HPFP (the one whose post-installation performance I've been reporting on). Specified temp range on these is -15F to 400F.

Bottom row: new "Chemical-Resistant Low-Temperature Fluorosilicone" from McMaster-Carr. Specified temp range on these is -80F to 400F. These were the only rubberized rings McMaster-Carr sold (or that I could find) with (1) a lower min temp than the Super-Resilient Chemical Resistant Viton and (2) a recommend by McMaster-Carr for contact with gasoline.



Above:

Top row: The new blue rings sitting on top of the old officially-unknown-material rings they are going to replace. Shown for visual reference. Size variances are minimal and almost certainly due to age/wear-use on the originals, though I noted that the original center ring from both the original HPFP and the junker HPFP seemed thinner (but matched each other). Wondering if this ring should be 3/64" rather than 1/16" wall diameter. If so, I have not seen a McMaster-Carr product that provides a 3/64" wall diameter with the desired materials spec and 3/8" inner diameter.

Bottom row: The same new Viton rings as in the previous photo, for visual comparison only. They are the same size as the corresponding new blue rings.

Anecdotal observation: the original blue ring has the same tactile proprties as the new blue rings. To the touch, the flex and hardness are identical. The original black rings are pretty hard (I don't know their age or mileage). The new black rings are harder than the blue rings (as expected -- the spec says this point-blank about the materials) and the new black rings are *almost* as hard as the old black rings. My suspicion is therefore that the original rings are indeed some flavor of Viton, and that the original blue ring is fluorosilicone.

I'm going to install these in the junker HPFP and swap the junker HPFP into service.

This is essentially a tweak to the original HPFP design, rather than an attempt to exactly replicate, if my guess is right about the original o-ring materials. Will advise as to results.

 

Side note, what program are you using on your phone?

 

OBD Fusion, on an iPhone.

 

Update: on the junker HPFP with the fluorosilicone rings. Vehicle started right up and went to proper PSI *immediately.* Vehicle behaved admirably until I got annoyed by a semi that threw a rock at the bonnet (mumblegrumble) and decided to take the opportunity to do a solid push on a slight uphill grade (and I do mean slight). I felt the car start to stumble at around 85MPH and the half-engine light came on. PSI immediately got knocked down into the 80s and stayed there for a good long time, and I finally pulled the car into a parking lot and restarted. I got a pending P0087 but the fuel pressure again came up immediately to normal range. It stayed there.

As I recall from my last pre-rebuild experience with this (the junker) HPFP in the car, it initially behaved well but finally got wonky after being driven around a bit, after getting an unexpected push, and then it was done playing in anything other than limp mode until restarting the car. So the behavior is consistent. The good news there is that it suggests the fluorosilicone rings on the solenoid are acceptable. The bad news is that it suggests that there is another variable in the mix. What is interesting is that each HPFP behaves in a way that seems familiar to me -- familiar to my prior experience with each *specific* HPFP, that is.

It may be time to crack open the other internals of the HPFP. I'm suspicious both of the condition of what's going on inside the thermal compensator (the other external housing) and of whether the amount/condition of oil in the HPFP itself is acceptable (since I've seen a "new" HPFP start to fail, in under 400 miles, after sporting a leak from that housing).

But in the meantime, is anyone willing/able to give this a whirl with fluorosilicone rings?

 

You Can use this for torque specs. https://www.allpar.com/fix/body/fasteners.html The last table has SAE into Aluminium.

Good luck with it. I have an R53 so have no useful info to add, but I am enjoying watching your testing!

James

 

Awesome link! Thank you!

 

Been doing some reading here on how the HPFP works, trying to get my hands on strongly technical discussions. Some quick takeaways:

(1) I agree with the folks who are pretty firm that the solenoid is where the action is at. Doesn't mean that other parts of the HPFP can't fail, but the solenoid is controlling fuel supply to the HPFP, period. No fuel supply or insufficient fuel supply and the HPFP becomes a really shiny brick.

(2) There is a post in this forum from 2012 that references hairline fractures within the piston pumping body. Unfortunately I can't find any follow-on. I haven't dug in hard with Google but anyway, FYI. This may be an issue for original HPFPs, if accurate.

(3) I am unable to figure out the role/place of the oil leak in my newest HPFP (the OSIAS pump from eBay/China) in its intermittent failure at less than 400 miles. I am simultaneously questioning whether the real issue with that HPFP may not be the HPFP, and may instead be the fuel rail pressure sensor as previously discussed. Or maybe I just need to fix its leak, plain and simple, as based on diagrams the oil is needed in driving those pistons (pressurizing the fuel). Hm, maybe I just answered my own question. Engineers, talk to me about what kind of oil might be inside that thing, pretty please?

(4) With the original HPFP that came on the car and the junker HPFP (which I suspect but cannot prove is also an original-manufacture), I am closing on the idea that if these things were lacking in oil due to a prior leak they wouldn't realistically be performing reliably at all under *any* circumstances. That takes me back to the solenoid and/or the fuel pressure sensor.

(5) Both my original and the junker HPFP are problematic with some shared trouble characteristics, but each also has its own unique patterns of misbehavior that carried forward from pre-o-ring-replacement to post-o-ring-replacement. As they are both using the same fuel rail pressure sensor (haven't replaced that) and share all other fuel system feeds and inputs, the only explanation I have for each HPFP behaving somewhat uniquely is that there is still something in each HPFP that is askew aside from whatever o-ring replacement would improve upon.

(6) I know the thermal compensator matters but I have no idea how it works and have no leads on examining *or* diagnosing it. Help?

I'm going to chew on this some more. But lacking a way to disassemble that solenoid, and lacking any input on how to improve or examine its behavior without disassembly, this may reduce to the next step being to replace the fuel rail sensor and putting the oil-leaking-but-youthful OSIAS HPFP back in the car, to see if problems go away. I don't like that tack, though, because a leaky HPFP is probably not a testworthy HPFP no matter what.

Would love to see some discussion here of this particular recap/analysis post of mine. I need engineering-driven ideas!