jonathan909

How is it for you up there, brother? We woke up to -40 this morning, and a few miles up the road (Sundre) it was -42 (not counting wind chill, of course). But we're due for a Chinook, so we'll be back up around 0C for a few days around xmas.

About the head bolts: Even though people in the business of selling them say that they have to be replaced, they don't. Re-use them unless they're obviously damaged (e.g. corroded threads).

I bought the Enginetech full-rebuild kit for my '99 EJ25 (DOHC) a few years back, and just for yuks, used the head gaskets supplied in the kit - I wanted to see how they'd stand up compared to the MLM gaskets we all know are the Right Thing. I got one year out of them before they leaked out, causing me to pull the heads and put in an MLM set. So the HGs in their kits are clearly crap, and I don't think I can recommend their head bolts either - some I got were, um, "less than straight". So the Enginetech kit is a perfectly reasonable - and very economical - kit, as long as you just toss the HGs and get HQ MLMs (incidentally, of late I've been buying Mahles from Rock - and they carry the Fuji Heavy Industries (Subaru) logo). Do it that way and you can get the quality where it matters (the HGs) and save a lot of money over a premium kit.

Yeah, that means that you'd get another data point, but it may not be definitive. The best test is the one you did, but the negative result could mean the problem is somewhere else (i.e. the actuator) or the two controllers are different. I would think a simple visual check (including comparing part numbers, if any) would answer the latter question; if they're different, trying the '99 in the '95 probably isn't going to work either. But don't forget what I mentioned previously - that others have reported simply reflowing (resoldering) all of the joints on the controller PCB and had it spring back to life. I haven't done one yet, but it's small and will only take a few minutes, so it doesn't make sense to me not to try.

Does that mean you've ruled out the controller as the possible source of the problem?

So problem in linkage or controller.

When powered from what? The controller, or on the bench?

I successfully used the external bolt biter extractors we discussed recently after nothing gripped the Allen head cavity (these are socket head cap screws).

Yabut... Removing the cover should still be considered nontrivial since there's a good chance that at least one of those 65 front cover screws is stripped. If that's the case, you're at least going to be pulling the rad to give you enough room to work at getting it out, and more if it gives you a bad time.

I have the same problem with the same model/year, only it's simply "reluctant" to move the door rather than absolutely stuck. So during the winter, once we get it to move to defrost+floor, I tell everyone just to stay the hell away from the control and don't ask for trouble. Why haven't I fixed it? Because I'd rather do three HG jobs and pull a Mustang motor for a timing kit (like I did this summer) than crawl under (or take apart) a dash once. But I know I'm going to have to deal with it sometime - probably this spring. Poking around online recently, though, I found discussions of the problem actually being with the control head PCB. Apparently (as with the instrument cluster) pulling it out and just reflowing (resoldering) all the joints generally solves the problem, so I plan to start there.

GD and I are at odds over this because he's practical and knows what he's doing, and I'm stubborn, cheap, and live by the old Stiff Records maxim: Try everything once except incest and folk dancing. I just did two HG jobs on the older ('01) H6 this past summer. Assuming yours is more-or-less the same (I haven't done a revised H6 yet), that furshlugginer timing cover alone adds a day to the job. Seriously, in addition to cleaning all that RTV, you're dealing with ~150 bolts, and that means it's all but certain you're going to get some stripped heads that'll add to the fun. I didn't change the chains, and so far it's been running great (one is in service, the other on standby and hasn't seen any actual miles yet). So I'm happy with the effort and the result, and actually had fun when it came to reinstalling the chains.

Rock: as long as you have no expectation of customer service, they're the best going! I've been perfectly happy with them since I figured that out. I think to date the only thing they've really gotten wrong was the timing cover crank seal (alone, not part of a kit) that they sold me for the '01 H6 - it was simply too small for the hole in the cover. Right part number, wrong fit. It was unclear where the mistake occurred - with the part's supplier or with Rock - and their error-bot credited me back half the cost of the part. Fine by me; not worth making a bigger fuss over a $5 seal that would cost $10 in postage to return. So, customer service reputation aside, do they not have anything that fits and works?

I didn't even know they'd rebooted it. If the premise is the same, I don't expect I'd feel any differently. MacGruber, on the other hand, is quite alright. Haven't seen the movie, just some of the SNL bits.

Hmm. I didn't see any mention of "running bubbles" in the coolant overflow, but is it worth doing a compression test? Hardly definitive, but maybe something turns up. I don't think I've ever head anyone mention "airhead" adapters on here - the shop air adapters that take the place of the spark plug and let you pressure up the cylinder without running or cranking.

Sure, but I was typing quickly. Don't want to get that OT, but worth noting that in its original usage, "hacking" was a hardware activity and only later came to be applied to the (now dominant) software usage.

I fuggin' HATE that word. There were plenty of words and phrases to describe this activity (kludge, hack, jury-rig, etc.) that predate by many years that idiotic POS TV show, and anyone who uses it within arm's length of me can expect to get bitch-slapped without delay.

I can't answer that one - I don't really know anything about the actuator and controls as I hate working in/under dashes so much that I've just been living with my '99 OBW's reluctance to change the airflow on command (e.g. from defrost to floor). In that case my suspicion is that the control PCB needs reflowing or replacement or something - I'm not having the kind of problem you are. Also, I don't know how representative a result you're going to get over <30 minutes. Myself, I'd pressure it up and leave it overnight. Consider that this leak may be (in part) a product of normal thermal expansion, and that it may be a smaller aperture when cold.

I'm a total tool queen, so if I can get even a cheapie to add to my permanent reserve (in this case it was a pressure tester from Princess Auto, which is sort of like our Harbor Freight), I'll do it regardless of how seldom I think I'll use it. It's about having as broad a range of tools in the arsenal as possible. Aside from the differences I mentioned above, consider that looking for the leak with the engine stopped and/or cold also means the fans won't be blowing drips, drops, and steam away from the actual source of the problem, and that you may be able to hear a hiss you wouldn't otherwise. An array of diagnostic tools means you're not relying on just one to contain all the necessary magic, as that's rarely the case. A different tool allows a different perspective on the problem.

Noting @azdave's caveat, it may not be the best way to go, but it's another way to go. The (obvious) problem is that there's only so much poking around you can do with a hot/running engine without risk of getting burned, your leg sucked into a spinning belt, etc. Looking while hot gives you the advantage of spotting steam, but a cold pressure test lets you get in closer. First see if it holds pressure. If not, the next questions are why and where.

Do you have a pressure tester?

Man, you're killing me. An old J-head with a motor drive on the table, and I'd be as happy as a baby in a barrel of tits. We don't need no steenkin' CNC. I did nine years (part-time) in a shop in SF where that mill was the centerpiece. I didn't get to use it much myself, and we didn't send anything into space, but the work we did blew minds (very weird story). I loved every minute I spent in front of it and can't wait to find the right one.

Assuming you mean "no coolant on the floor of the garage", how about on the passenger side floor pan i.e. a leaky heater core?

But really great milling machines.

Yes, I actually meant that - and while we're not Winnipeg aka Winterpeg, we do get spells of -40, where C and F converge. That's why we're so pleased with Subi performance overall in that respect - with a not-terrible battery in the car, we can usually start down to -30C without plugging in the block heater. It's not happy about it, but it starts. Trivia: I'm from a town a couple of hours south of here called Pincher Creek - it's just north of the Montana border. This is the sort of thing we get in the winter - record-setting slew rates: https://globalnews.ca/news/4835093/temperature-change-swing-canada-january-10-pincher-creek/

That's helpful. Can you offer a seat-of-the-pants estimate of the mileage on an engine that would begin to qualify it for the heavier oil? Aside from my just being a stingy and sloppy sumbitch who's fine with whatever 5W30 or 10W30 is cheapest on any given day, I'd be concerned about the heavier oil making very cold weather (-30C and below) starting a lot harder. One of the terrific features of these engines is that they're easy to start consistently at low temperatures without plugging in, a far cry from the Detroit iron 8s of my misspent yout.