Rep Power: 0

So, do you guys follow what I'm saying? That the speed at which the input shaft is turning when you shift, affects shifting even though the crank is disconnected at this point? Remember too that the wheels, axle and diff are connected to the output shaft, and when the car is off-power it is also decelerating. So the slower the input shaft is turning at the point of disengagement, the more synchronization the cones have to do to match gearset speeds?

Imagine a fan spinning ~120 rpm, and you with a paintball gun. Is it easier to hit a blade of the fan if it's spinning with very little variation, or if it accelerates for 90 deg, decels for 90 deg, etc.? The mass of the input shaft demands the synchros be given very consistent speeds/accelerations to work best... and does its job at low rpm best with as little variation in rotational speed as possible... which is when most drivers experience shift balkiness -- low rpm, stopped or moving very slowly, or short-shifting. Friction due to pumping loss or oil degradation can affect this, as well as possibly brake drag or a misbalanced wheel can do it from the output shaft side.

I really don't want to confuse too many on the issue -- just trying to explain what my view is on what's happening. Again, if you don't agree with my reccie, I have no problem with that -- I don't have any agenda. Seek your own solutions.

Here's a demonstration of the crank acceleration phenomena in an inline-4 (and Yamaha's excellent, if highly application-specific, solution for it, @ 2:28):

 
Rep Power: 518

In a car:
The input shaft can only be spinning as fast as the previously engaged gears' ratio allowed --- regardless of the engine speed.

Clutch disengaged: Engine is disconnected and has no bearing on input shaft speed now.

As soon as you pop the shift lever into neutral on the way toward the next desired gear, the input shaft begins to slow down. Wait a sufficient amount of time for the shaft speed to match the RPM of the next upshift gear, and you won't make the synchros work near as hard.


The inertia of the spinning input shaft comes from the rotating mass of the clutch disc AND the spinning shaft and gears inside the trans.



The cones of the synchro rings can be thought of as simple brakes used to help match speeds between shaft/gear/hub prior to actual tooth engagement by the sliding hub.



Interesting. Looks like it should feel very uneven, moreso at low RPM/high load.
But I suppose at 14k RPM you wouldn't notice the uneven feeling much.


Back to regular car stuff now?

 
Rep Power: 0

Yes yes, I know all of that. I also know my driving style (when I move the shifter after letting off the gas [quickly], and how fast I move the shift lever from one ratio to another [quickly]) hasn't changed at all from before vs. after the AF + oil change. Shifts much better (it actually balks more if I wait too long to move the shifter from NEU into gear). The only thing I could find in the Civic manual powertrain that could logically account for why, was what I posted.

I even tried double-clutching before the AF/oil change, to see if that improved shifts. Nope (and man is that a lot of work for one ratio change)... same thing, if I wait too long and the flywheel slows to idle speed, shifts get rougher than if I just shift AND declutch AND move the shifter into the next ratio, as close together as possible. Then it feels silky...



Have you tried riding one? I have, when they first came out in 2009...

The R1 requires a balance shaft at low rpm due to it being solidly mounted in the spar chassis, but no more than a Honda V4 would at the same rpm, which is super-smooth and torquey, just with a bit more character at idle and just above (and this I-4 architecture is waaaay more space-efficient). It feels eerily similar to the VFRs when leaned over getting on the power... glued at the back wheel, even when it breaks loose (which you need to try pretty hard to do, or just get sloppy... which isn't an option for me on a 180-hp literbike). In a car it wouldn't need one, just rubber mounts as is the usual way. However, the uneven power pulses below ~4000 rpm would be hell on alternators, power steering pumps and esp timing and acc'y belts, which is why there's no reason to spend the money to adapt them for an automobile (which make no mistake, would be pure exotic to hear rev through the gears... if no more fuel efficient or powerful than a much-cheaper-to-make conventional Four).


Imagine that, but with double the displacement, in 4G63 form (the easiest one to convert to crossplane, as it already uses a balance shaft)... woof. :P


See, that sounds like what most expect an Inline-Four to sound like. Much higher pitched. A crossplane sounds and puts down power exactly like a 180-deg crank V4 (Honda's 3rd-gen-to-present VFRs), with the packaging of an Inline-Four.


Man, makes me miss my VFR yet again. :'P

Okay here is pretty much *exactly* the bike I had, right down to the Staintune exhaust (that Staintune was the only exhaust that should be allowed on the 5th-gen VFRs by law ) See how much like the crossplane it sounds? That lope-y, V8 sound is due to 90-degree-phased combustion -- though on a VFR it's the angle of the heads on a flatplane crank that makes the lope, vs. the opposite on the R1.


Never left. Not regular car *talk* though. But it was just a reccie on an easy way to improve shift feel. In the interest of just moving on though, I'll close my comments on this subject here.

 
Rep Power: 0

Real fast... checked the usual coolant tank level and clutch res levels this AM... and found my coolant level's actually falling now, instead of rising. Great.

Actually, that is great... if it isn't the coolant leaking into the head and being burnt, then there's a leak somewhere. That is muy preferable to a looming head gasket failure...

I have no access to the car's service records, as I have no idea when it was shipped from Canada (probably the PNW more likely). If the head gasket was done already... then I hope it's done and done, and I'm just dealing with a rubber hose problem.

If not... I may become my own test case re: how long a replaced head gasket lasts in the D17A2.

And ezone... re: ARP TTY head bolts for the D17: no answer after two emails. I'm going to assume I'm being blown off at this point. Too bad it's with lots of teeth.

 
Rep Power: 518

Note ambient temps too. Res level can fall with ambient temp drops. Wait, I bet temps are pretty stable there in Hi.

Built in Canada? Yes, plenty of US market cars were and are built in Canada.

If you didn't ship it to Hi yourself, I'd assume it was sold there in Hi. brand new. A dealer VIN report would reveal the original selling dealership.
Too bad. I wonder if someone else handles TTY bolts in the size needed for this engine. Victor Reinz? Fel-pro? Gasmiser? Dana?



Never been a sportbike kinda person.

 
Rep Power: 0

^ Thanks, will do...

RE: 2HG... the seller told me the lady he bought it from was from the Pacific Northwest (PNW), probably further inland since it's surprisingly free of rust it'd have, were it sold new here on island. I'd guess Oregon, since a *ton* or Oregon cars get sold there then shipped to HI (cheaper by thousands of dollars, depending on what make/model).

RE: TTY: Applications for this fastener can't be THAT rare... but I don't have a clue what the requirements are for sizing TTY bolts, as it's the twist of the shaft that provides the clamping force through temp cycles, not the threads... so overall length may be more crucial. The ones I've worked with on Bombardier 2-ST engines looked like those twisty tall ballroom candles in design. I'm sure there HAS to be an existing bolt from another vehicle that will work, but finding it is the key.

Since the D17 doesn't spec TTY from the factory, I doubt calling an aftermarket OEM supplier will get much traction (hell, ARP, a pretty good speed part prospect, is radio-silent about it). Maybe some racers that specialize in D17s? I haven't the foggiest if D16s are similar enough, as their engines and owners would actually buy and support speed parts made for 'em... but if not it's square one.

RE: riding the 2009 crossplane-crank R1: they're old bikes now... maybe you could log some miles on one sometime (yeah right ). Never too late to succumb to the sirens of agility, acceleration and braking, though... I didn't start racing myself until I was 36, but had wrenched on the other side of the pit wall for a lot longer.

If you ever get the chance to ride a non-crossplane and the crossplane R1 back-to-back... it's astounding how much more traction the newer bikes have mid-corner, getting on the gas. Made me wish my VFR was about a hundred pounds lighter, as they feel very similar, except the R1 had tons more front end grip as well (inline-4 vs. V4 weight distribution).

 
Rep Power: 0

ezone was right about the coolant tank level... though it's quite a bit more than any case I'd seen in cars or bikes thus far...

So, I tend to check my tank level every AM about the same time, betw 9am and 10, day is still pretty cool, about 71-73 degF this time of year. Tank level was slowly filling after purchase when I topped it off (seller sold it with nothing in the tank, great). After several weekends of burping, topping or removing coolant, rinse & repeat... it finally stopped overfilling a couple of weeks ago, finally...

However, most of the cooling systems on motos and cars I've worked on don't push as much fluid past the rad cap as this car. I'm seeing 2" over the MAX line in the tank... not sure if that's normal, but it is drawing it back in when fully cold (back to MAX line). Certainly, if the motor were running the full 1.1 bar/16 psi of pressure during load... there still wouldn't be THAT much overflow with the temp gauge reading fine. I am getting good mileage, but not the best consistency with power as the day gets warm... hmm.

Used to beat the living crap out of the Paseo, and never went a drop above the MAX line (actually moved from Phx to Seattle with my car packed with hundreds of pounds of junk, over mountain passes, without a single level change).

I'm thinking, why not just go to a stiffer cap spec (1.3 bar/19 psi)... but then remember the OEM crimped plastic tanks and spring-type hose clamps all over the motor. Don't want to trade a few days of steady tank level... for a strand. Being stranded with a cracked rad tank or spitting hose joint would suck. Maybe the cap change could only happen with a Mizu rad at the very least...

 
Rep Power: 518

Sounds OK to me, it's doing what it's supposed to do.

If you have the level set in the narrow part of the res tank, it will rise (what appears to be) a lot. If you have the level set in a wide part of the tank, the distance it rises would be little.



Higher cap spec would raise the boil point, and it's not necessary. And wasn't designed for it.

 
Rep Power: 0

The plastic rad tanks are probably fine for a mere +3 psi of system pressure... but I'm also dealing with an unknown quantity, not to mention a 10-yo rad. Brand-new the stock rad and hoses'd probably do fine with 19 psi. No idea if the coolant is fresh or not, which would also affect cooling performance. I'm gonna assume given the rest of the car though, that it too is shot...

Hmm... I'd never heard of a system that purged past the pressure spec regularly... but then again I'd never seen an N/2/B-type cap, either (even my motos all had S/1/A-type caps). So be it.

 
Rep Power: 518

They don't.

They reach the max spec pressure (cap rating), then the cap vents the excess volume to the reservoir to maintain/limit the radiator at max spec pressure.

 
Rep Power: 0

Very odd (though really, that's what I said. ). Never seen even 120-hp/liter 600 I-4s do this... and I've seen a lot of hot 600s on my lift (not to mention their small Dali-esque coolant tanks). Cooling to ambient, they don't seem to draw as much coolant to lower the level... maybe it's just an optical illusion with the D17's narrow tank neck. Even in the narrow part of the tank though (at MAX line)... that's gotta be a cup of fluid coming through that cap when warm, and that's a new one, at that... weird, but apparently just to me.

Since some built D17s seem to keep the same spring clamps and hoses, think I'll keep those. But before I think about a 1.3 bar cap and Mizu all-alum rad, I'll see if a full coolant drop & swap will help the dwindling response when hot (just took a drive to a new coffee shop upcountry on the slopes of Mt. Haleakala, and it couldn't get out of its own way the whole way up (drank a ton of fuel as well -- took about 20 miles off my 1/2 tank range )... so wondering if the knock sensor is removing a lot of power. This mill was genuinely peppy (and then some) with a CEL from the wrong sensor in the upstream location... almost tempted to swap the damned thing back in and eat my loss of 4 mpg combined (30 mpg).

 
Rep Power: 367

i have not being following, since you two are weathered honda guys, but what the hella going on in this thread? am getting lost :P

 
Rep Power: 0

I'm just documenting issues related to driveability with my '04 EX manual, since it's the first Honda (car; my VFR was the first Honda ) with my name on the title.

The best thing about this thread, is ezone, an experienced car and Honda car tech, can pick apart my assumptions with them due to my moto-only wrenching experience. I'm learning metric tons of data... though he may shake his head so often reading, that it may be rotating 360 degrees around by now.

But I *have* to know why the car versions of components are so different (esp symptom-wise) than the moto versions, because that way I can come to so many of my own conclusions when stuff happens. That's why I have to challenge him with questions, or less that, do my own tests based on our and other poster's convos here. Ezone can't possibly know the answer to all my darned questions (esp since both of us have some engineering knowledge... the tangents can go pear-shaped real quick )... but really it's all down to understanding the source of problems, to pass tips and advice on that's worth something.

Let me know if there are issues you'd like addressed in the thread if any, and I'll make it happen.

 
Rep Power: 367

nope, just admiring a bright new member acquisition to our forums thanks!

 
Rep Power: 518

Does it show?

You can't defeat the laws of Physics.

X amount of liquid will expand Y volume with Z rise in temperature.


(Imagine the system constructions and operating temps are exactly the same here) Car cooling system holds more liquid than a bike (maybe more than double?)
Therefore, for the same rise in temp, the car will have more liquid be displaced into the reservoir than the bike.

The rubber hoses can "hide" some of that due to ballooning, it takes volume to make them balloon. Air pockets, if any,will also hide some of that expansion.


Yes, the narrow part of the tank can cause you illusions.

If the systems used an airspace in a pressurized expansion tank, you probably wouldn't ever notice any of this.



=

I overate on pizza long after dinner hour. I'm paying for it now in the middle of the night.

 
Rep Power: 0

It's obvious. Though that statement pretty much proves it. I'm quite grateful for your patience for the aforementioned reasons.



It's possible that in a bike there may be spots in the system that are virtually impossible to burp the air from (V4/V-Twins esp). This may be accounted for by design, as I've never had a customer comeback for *worse* cooling after I've serviced their vehicle... this in hotter-than-two-rats-****ing-in-a-wool-sock-Phoenix. The usual burping, followed by a cooldown and gentle rock of the bike to release trapped air, rinse/repeat until no more level changes in the rad. Check overflow level, test ride. Zero change in tank level. Take money and next RO...

There's more coolant, sure, but there's also a physically much bigger coolant jacket (shoot, the D17 long block is more than twice the size of the Yamaha R6 engine AND trans), not to mention the coolant tanks are similarly larger on the D17. There's also the issue of a heater core. By that logic, if the proportional size-up is the same, then the change in level should be proportionally the same... even in favor of the D17 when the heater core circuit is open. It isn't.

What I think could account more for the discrepancy... is the capacity per engine size is a lot higher on a sportbike vs. the D17. 4 qts/1.7L vs. 2.5 qts/599cc on a 2004 Yamaha R6... the D is almost triple the displacement (easily more than 2x the physical size) without having twice the coolant capacity. I'd have to check... but I think the discrepancy isn't the same with a B17... which has the potential to create a lot more heat, thus needs more coolant to make it reliable. Honda probably spec'd right on the edge of adequate for the D17s... which sounds familiar.


When I did a yearlong gig in graphic design in Phx ten years ago (right after leaving dealer teching and concurrent with my indy shop startup)... I did my absolute best creative work betw 10pm and 3am. Used to walk in the middle of the night along the canal to this QuikTrip to get a bucket-with-a-handle of Diet Coke with Lime... and three The Bomb burritos, which would be my art fuel. Add an 'f' to that, and that would describe 4am - 9pm the next day.

Ah, being young and ignorant of your health. Changes when you realize you have fewer years ahead of you than behind you.

 
Rep Power: 518

Yeah, they can overheat with only a quart missing. They do run right on the ragged edge of cooling capacity, no reserve, no room for error.


Ignore engine displacement. Ignore water jackets and heater cores. ALL irrelevant. Only think of heating up water.
The engines burn fuel to heat up the water.


You have 2 gallons of water on the stove, container is completely full and sealed, with a skinny tube sticking out the top to measure expansion of the liquid...... You heat that water from room temp 70*F up to 195*F and measure the volume of expansion that went into the tube.
Maybe...6 ounces of fluid displaced. Tube is really skinny and 6oz covers a lot of distance.

On the other stove burner you have 2 quarts of water setup the same as the first. Repeat the heating and measuring test. 1.5oz of fluid displaced into the tube. This tube might be fat so you don't notice the change much.

Um, I eat enough Subway to cover this one meal of indiscretion!

This was a fluke. I hadn't had real pizza in months. I probably won't have it again for several more months now.

 
Rep Power: 0

Yup, know all of the basics. But when you have a 75-hp/liter 1.7 heating up 1.0 gal of water, and the same engine heating up 0.7 gals of water, which do you think will expand more? It's the same reason it takes longer on the HI setting on your burner, to heat up a gallon of water vs. 3 qts of water -- the smaller amount will expand more, because there is nowhere to go but the overflow... and if *that's* biased towards showing more change vs. a Corolla or domestic... it'll alarm people.

The sportbike has less capacity for absorbing heat... but a lot more rad surface area per volume than the D17... which is why it can shed its heat faster. Useful when operating at 15K rpm at full load with a 200-lb passenger.


You're making the assumption I was criticizing you ... when I was only talking about my own gluttonous youth... nothing was aimed at you personally... are you kidding me? It's like having my own Pro Honda class comin' on here!

BTW I eat Subway every day, without exception... the same sandwich... and have lost and kept off 50 lbs since last May. Wise man once said... to one person you may be the world.

 
Rep Power: 518

The expansion is exactly proportional.

Given all other factors as equal (amount of temperature change): If one gallon expands by 3 fluid ounces when heated, then 2 gallons will expand 6oz. A half gallon will expand 1.5 oz.

1 fluid ounce = 29.5cc

I'm just tossing numbers here, I don't know the actual amount of change off the top of my head.

 
Rep Power: 0

So... just experienced a weird twist on the driveability issues...

Was getting on it on a small local set of twisty technical roads... when my temp gauge went cold, still at speed. Pulled over shut down... nothing bad underhood, coolant at the same old levels as always. So on a hunch just decided to check all the sensors on the water outlet (where the t-stat is). Just pressed in and jiggled all the weatherproof connectors... when one clicked. It was the ECT connector.

I thought then about some of the issues that plagued my 15-yo Escort, and one of them was loose or corroded connections on the circuits the PCM needed to determine fueling -- CPS, CamPS, ECT, MAF (MAP on speed/density systems like the D17s?), IAT, ambient air pressure (not sure what the acronym for that is). Once those were fixed, the car ran much better and didn't feel like it was struggling just to drive normally. Pro tip: don't use a K&N oiled gauze filter on a car with a MAF sensor.

Anyhow, got back in and started the car -- got my temp gauge back. Not only that... but a *lot* more power. I mean, a LOT -- almost as potent as the car was with the CEL and wrong upstream sensor (almost). This with the car piping hot, where it'd start to get meek in the past. Now it felt like it did right after warming the car up in the AM! Even shifting was markedly better, esp on downshifts...

My past comments about shift quality only make sense when it's assumed all other functions are happening as designed... but this makes much more sense as a cause for weird driveability -- if there was an intermittent open in that connector, my guess is the ETC'd either be telling the PCM to change fueling constantly according to wildly fluctuating conditions... or it'd be resetting maps every time there was an open, and the PCM would have to start all over again from baseline. Just tried a simple test... how fast can I get to 40, when I attempt a WOT start? Well, let's say the car surprised me. A lot.

One click... lesson learned. Even Hondas can develop intermittent opens on major sensors after ten years. *sigh*

 
Rep Power: 518

BARO

Yes there is only one ECT, the computer uses it and sends that info to the gauge.

I'd think a human left it loose though. They don't normally come unplugged on their own.

 
Rep Power: 0

Ah, thanks... noted.


Oh no, I meant check all the connections on the casting, including the ECT. There's a rad fan switch there IIRC, checked but it didn't click.


Could be. I had to remove the ECT connector to get to the cam plug and again for the valve cover gasket swap, so maybe that's when it happened. Pretty difficult to *not* feel the click though (and I'm more than experienced enough to feel for it)... so maybe I'd jostled it too hard somehow when putting the harness with the coils back on & just got unlucky (never known one of them to pull loose... but then again this connector's seen 10 yrs of being knocked around, right above a very hot area).

I'd had a similar problem on my Escort, but for the knock sensor and injectors (both similar two-wire pin-type weatherproof connectors) after finding someone hadn't utilized a malleable wire tie for the knock one, and it'd been whipping around with the ghastly-rough-shaking SPI motor at idle for 15 years. The injector ones had click-barbs that came apart in my hand when trying to remove and inspect them ... which is why I feel it may be either reason.

But if not, I'll gladly take the hit, as my power and driveability have greatly improved -- the car moves on A/C now as quickly as w/o A/C before. Just drove it another fifteen minutes in town and on the highway at speed, and it's a different vehicle. Accelerator pedal feels 'hard', like it's wanting to accelerate now from steady throttle settings, vs. meek and spongy response but smooth before. Something's definitely changed for the better.

 
Rep Power: 0

Interesting... then why does a D17 cooling system less a quart of fluid overheat out your rad cap? It should expand a smaller amount, by that logic.

It's because a volume of coolant has a thermal reserve, that is, the amount of heat it can absorb before sublimating into steam. Approaching this thermal reserve (with less fluid in the system), the fluid will expand more than the same system when full, as there are more coolant molecules to share the absorption of heat. The full system simply divides the heat load by the fluid capacity, while the short one has to divide the same heat among fewer molecules -- exciting each one up more, as heat does. More excitation = more expansion, and more fluid past the 16 psi limit on the cap, ending up with more fluid in the tank.

A lower volume will also affect the amount the rad can cool it, since there is less fluid to touch each fin with a quart out than full. It can't take full advantage of airflow to remove heat, as the fluid isn't always constantly filling all available fins. Air is a terrible conductor of heat compared to coolant. Thus, even more expansion as the moved fluid doesn't shed as much heat before it touches the head/cylinders again.

As close as Honda shaved the thermal reserve in that system to match the heat load of the D17 head and cylinders (probably to save cost and weight) ... it's no wonder there's a lot of expansion. It may be normal... I've just never seen a cooling system fill up a tank that far over MAX warm, when it's filled to MAX cold.

Oh BTW -- just finished driving about another 20 mi with the car with properly-fastened ECT connector , in rapidly cooling conditions from ~4pm to 8pm. Feels exactly the same as during the day -- there used to be a *huge* difference in character once the day got warm, then cooled in the evening. Now there's zero -- which tells me perhaps the PCM was being told by the intermittent open on the ECT to enrichen even more than the super-rich OEM mapping Hondas already use -- that'd explain my smooshy response up to now in the heat of the day, but cooler temps making it behave better.

 
Rep Power: 367

glad it's OK now.
like they say in TN, "click it or ticket" (for seat belts :P)

 
Rep Power: 0

Hehe -- yeah, I'm pretty careful when working with esp electrical components... but vets aren't any more perfect than anyone else... we've just had a lot more time having that proven to us, over and over again.

The great part is... this problem explains a lot of the other little things that were happening, as well. If the ECT was intermittently opening, then the richness that resulted also explained why the car seemed to climb grades so smoothly, yet got awful mileage doing so (you need more richness for greater load) -- the BARO was just trying to tell the PCM to lean out, but ECT will take priority... and if it was being interpreted as 'cold', then that's what was being injected.

Just went upcountry today to have lunch at a café up there, over the same road, within an hour of the same time... and the car was waaay more responsive, always feeling like it had a bit more in reserve... whereas before it felt smooth but winded even climbing slowly.

I'm also not seeing the fuel needle budge from filling it yesterday, even including that climb and the run back down to sea level... 45 miles, and the needle hasn't gotten into the indicated part of the tank yet. It also doesn't feel as eager to run from cold+1-min-warmup like it used to, also a sign that mixtures may be a lot leaner.

 
Rep Power: 518

If you overheat, you are way beyond the 180*-212* normal operating temp range.

I wasn't dealing with that in the stove/expansion example.
You lost that reserve when you lost the quart or more from the radiator.


You mean boiling?

You're still way over normal operating temps at this point. You can overheat without boiling too.


==============
Here's a goofy question for you:

Does an average high revving bike engine use ordinary ethylene glycol, or does it have special conditioners, or completely different coolant types to deal with the ultra high RPMs their water pumps would be capable of turning?

I could see cavitation as an issue at high RPM, just wonder how they deal with it.








You still seem to be looking at rise (length measurement) without considering the diameter of that skinny neck (volume measurement) at that part of the bottle.

I'm guessing.
When the ECT circuit is open, the PCM substitutes a value that will (supposedly, under most conditions) still allow the car to start and run, even though a scanner datalist will show the temp as -40*F.... I'm not sure if it is programmed with a fixed substitution value, if it is a variable value inferred by using the IAT and other sensor info at startup. Probably preprogrammed fixed value on that PCM.

I have never seen actual documentation on the failure strategies or programming of their PCM.

You could probably SWAG at what it does by watching AF and STFT response to fueling changes as you disconnect and reconnect the temp sensor while running at various temps.


You should have had a CEL on with probably a P0118 code if the ECT was open circuit for longer than 2 seconds.

 
Rep Power: 0

That was my point. This is why there's a gallon spec'd for the D17, and not a quart less -- you reach the thermal absorption limit of the water, and excite its molecules so much that they phase-change into vapor (steam), esp when other problems allow pressure bleed to occur below spec.


Yup.


As above, adding pressure to the cooling system will raise its boiling temp... but if the cap isn't spec'd to handle more pressure, it will still be the point of failure (by design). However then the no-boil overheat only reinforces my point -- you can still push more fluid into the overflow tank due to a smaller volume of coolant reaching higher temperatures, than a larger volume of fluid. As long as engine temps are more than capable of boiling over that volume of coolant without being constantly cooled through heat exchange thru the rad, the thermal reserve model still holds.

If you were to create a force field around a teaspoon of water that would pressurize it constantly above the boiling point, and expose it to a burner plate... its percentage of expansion would be higher than the same situ with a cup of water under the same conditions, on the same burner. There is less water to absorb heat, therefore the teaspoon will expand beyond the cup, if you divided it into teaspoons and measured each.


They do, kinda -- even back when Yamaha introduced the first 15.5K, 600 sportbike with over 100hp stock ('99 R6)... they spec'd plain ol' Prestone 50/50 or water with Wetter in it as fine for their systems... nowadays even the 17K race R6s still use water with a surface-tension-collapsing additive -- no problem.

The secret with those motors... and kind of a segueway back to our discussion... is through more volume. The pump doesn't have to necessarily have to spin fast enough to cavitate, as long as it moves a large enough volume per revolution of the crank to satisfy requirements. In fact if you look at an R6 pump, they're very crude designs compared to impeller designs in watercraft:


On the cavitation issue... all that's needed is to underdrive the gearing on the pump, to prevent it going over the threshold where the water'd start beating things up. Often these days they just use a beveled shaft to key it to one of the starter's reduction gears (as on bikes the starter is geared to the crank and never disengages as on a car). And on race motors... you can bet cavitation exists... but as long as enough coolant gets through the rad? Who cares, that's why they sell spares.

If you look at the D17 rad and the size of its head and cyls, the size ratio between it and an R6 isn't that different -- but remember that surface area only squares as volume cubes. So in reality you can tune a 600cc I-4 with much, much higher compression for hp/liter and still adequately cool it with a given volume of coolant. Engineers aren't naïve though, so much more margin is designed into I-4's to deal with the fact that isn't as much metal there to absorb heat as enjoyed in a typical modern car engine (you will rarely see a non-racecar run at 75 - 100% its redline, 45 mins at a time, for its entire service life). So Yamaha and other bike marques use very deep cores to make sure there's a large drop in temp across the rad.

An interesting effect of cooling on the moto side, is that liquid-cooled V-Twins tend to need much smaller rads for a given displacement. This is because on a V-Twin, there are four sides of the cylinder exposed to incidental airflow. In a I-4, there are none -- two cyls with three sides exposed, and two with only two. That's why 600s need the same size rads as a 1000cc V-Twin like the SV1000 or TL1000R. Incidental airflow does count... which is why there's so much crap on cars now to manage it.

I want to get a Mizu rad, for its increase in that temp differential as well as more volume, insures cooler temps under all conditions, thus more margin away from failure. They seem to accept all the stock parts such as both stock fans, as well as drop into the stock holder, plus give the added safety of welded tanks. I've heard of too many crimped tanks going plooey on the Escort after the 10-yr mark + chintzy aftermarket ones. Cooler head/cyls also lower oil temp a bit, so bonus for durability there.


Well, based on a previous post describing the weird (Dali-esque) shapes that factories use to stuff a quart bottle into the fairings of various sportbikes... I'm familiar with it. Even so, 2" above the MAX line, is a third of the distance between MAX and MIN. That's not a ratio I've seen before, even taking the gooseneck into consideration. So should I fill to 2" below the line cold, and expect to check it at MAX when hot? Is that how Honda says to fill and check it?


That's what I was thinking too -- why didn't I get a CEL on my dash? Of all the conditions that the PCM uses to insure proper metering the ECT signal is one of the most crucial. I don't seem to have any pending codes on my look through the Torque app either. Have no idea what Honda's PCM strategy is for dealing with an intermittent open on the ECT circuit... but all I know for sure is... I am quite pleased to no longer have one (as of now I have 57 miles on this tank, and the needle has just begun to move -- with far more power and response. It was ~33 - 36 before.

 
Rep Power: 518

Operating temp is below the boiling point.


So they don't have cooling issues if the engine idles too long, since the water pump would be turning so slowly?




Fill to max line cold, there is plenty of room for normal expansion above that line. Check again after cooldown, level should be where you started.

During any service, I fill them to the max whenever I have to fill one. No issues.
Sometimes I miss and get too much in there, it's really hard to see sometimes.
A little bit overfull goes quite a ways up that narrow part of the tube.
What's your owners manual say?

I can't answer why no code for you...... My info says the threshold is 2 seconds @ sensor output above 4.92 volts, and it's a one-trip code set.
Maybe it wasn't truly open circuit?

Maybe your app can't pick out that code? (Try that yourself, disconnect sensor to force the code to set)

 
Rep Power: 0

Missing the point. Given a heat source with enough heat to boil the larger volume completely to vapor, a smaller volume of coolant will undergo expansion more than a larger volume -- the thermal reserve model in a cooling system with a heat exchanger (rad) still stands. Heat has nowhere to go, but to excitation of the coolant, so even though in our imaginary cooling system it can never boil... it will certainly expand, right into the coolant overflow tank.


Missed the part about greater volume in a moto I-4 cooling system from earlier -- need to read through the posts a bit more. More volume through a bigger core is enough to cool a modern bike I-4, though their fans come on a lot more than the same displacement liquid-cooled V-Twin. They're turning pretty fast through gearing at idle already... but they're sized so they don't have to spin that fast to move the required amount of coolant. The R6 one is the same size as the D17 one, if not a bit bigger. My SV one is also bigger than your D17 one, come to think (~5" x 4", with s 2" diam x 3/8" impeller, for 650cc).

Cavitation at redline is definitely present I'd imagine, with that stamped-sheet origami business -- esp compared to top marine race drive impellers. But then again, in racing that's why you buy spares. It doesn't seem to affect the bikes achieving consistent output at race pace at all, so not a concern.


Okay -- well if that's normal, then it's normal. It should be stressed that this rising that much above the MAX line shouldn't be confused with the dreaded HG problem... but if one sees the level continue to rise beyond 2" hot... then that shold be the sign to prepare for a swap.


You don't know? *bends ezone's tech dogtags*


No prob -- it may just be Torque as mentioned -- test with disconnecting ECT is easy enough, will do. If it was open, it wasn't for 2 secs, so maybe just vibrating open? No idea... but man is it fixed now. I do want to get a dedicated code reader, though... you know what they say about a Swiss Army knife... does lots of things, none of them particularly well.

Just saw 57 miles on a full tank, and it wasn't even registering on the gauge yet -- that with another climb up the mountain and back this AM. Totally different car. And yes, my steering feel and shifting have also miraculously improved. Maybe the extra power is causing a bit more toe-in due to play in the LCA pivots now? I still need to align the front end... but 210 miles on a half tank with the bad ECT connection ... alignment isn't that bad.

 
Rep Power: 0

Just tested -- yup, P0118. Looked it up and it's a High Input code, meaning there's too much resistance, and an open is infinite resistance, so makes sense. However Torque can read pending codes and none were saved to be cleared. It reads codes on the fly but you can't take a snapshot to study codes set over a given time, or any kind of scoping function, which a decent code reader could show me. Could also be the BT code dongle I have... it's good but there're much better out there.

One negative after finding and correcting the ECT connector issue... is that suddenly now I have trouble getting into first after idling in NEU for a bit at a light. That's never happened before, even when the car was still aching for some love at point of sale. All other gears seem to shift much better now and clutch action's gotten way better as mentioned before... but this is new. Can sometimes even balk with my clutch to the wood, and a shift into second or fourth before selecting first, even. Don't like to sit with the clutch disengaged as my slave cylinder may need a rebuild, so idle in neutral. We'll see if it improves with a trans fluid swap this weekend (seller said it was done along with the timing belt, but who knows).