Each new 13B-REW is handbuilt by a single engineer
 

I found this interesting article on the Mazda site that describes the rotary engine manufacturing facility at Mazda.

Thank you engineers of engine group number 3 Tasuya Happo, Koichi Iwata and Yoshiteru Inoue for hand crafting our engines.

And now you know who Vargas bros are slandering when they make fake videos showing impossible "assembly errors" on new Mazda 13B-REWs.


https://www.mazda.com/en/innovation/...story-2021-04/

the most exciting part of that article was being able to see some of the super OEM stuff they use to assemble the engines. that clamp that holds the motor together while assembling it and the cradle thing that holds a fully assembled rotor..... super cool stuff. id like to see them do an engine from 0 to packaged. just seeing how mazda does what so many all over the world do their own particular way is just such beautiful insight.

We have installed two so far and they are awesome.
In our opinion the only way to replace an engine unless you enjoy trouble.

That video seemed fishy to me. They made a big deal talking about not cutting the camera away when they started disassembling the engine, then they cut the camera away just a few minutes before showing a major problem.

https://www.rx7club.com/west-rx-7-fo...-shop-1133057/
https://www.rx7club.com/bad-fugly-bu...ports-1148511/
https://www.rx7club.com/3rd-generati...3brew-1144484/

Some things I saw watching the Vargas video.

They removed the flywheel nut (hmmn, a cut scene...), removed the flywheel (not shown) and then remembered to check eccentric shaft end float and put the flywheel back on loosely ensuring an incorrect end float.

How do I know this?
Because in the subsequent scene he whacks the flywheel lightly with a deadblow and removes it.
I wish it was that easy. When you remove the fully seated flywheel you have to put the flywheel puller on, heat the flywheel, tighten the puller and whack it, tighten the puller and whack it till ... BANG! it pops off and slams against the loosened flywheel nut startling the fuck out of you everytime.

Intererestingly, I did measure the end float on my brand new crate 1999+ high power spec part # Mazda crate 13B-REW. I Was surprised to find it was .0039" when the 1993-1995 limit is 0.0035".

I looked at RX-8 spec and its 0.0039" so, I ran it. I also ran it at 9,000rpm because I had read the 1999+ high power spec had factory side cut rotors.

Teardown after 20,000 miles and 3 race seasons did reveal the side cut rotors and no abnormal wear or damage from frequent 9,000rpm operation or 0.0039" end float.

I havent been able to find the 1999+ jdm manual to confirm if there is an official revised end float spec like on the RX-8.
______

The "bent" rotor oil bypass thermal pellet used as the video thumbnail and attributed to metal in the engine.

Just before the cut scene you can see the thermal pellet poking up out of the eccentric shaft.

When I saw the bent thermal pellet I wondered "would that even fit into the eccentric shaft?"
The OD of the pellet seals against the ID of the eccentric shaft to block oil flow when heated to closed position. Not a lot of room to be cocked askew by the out of alignment disk at the end of the plunger shaft.

I paused the video and bent a thermal pellet at less angle than shown and put it in the eccentric shaft.

It went into the eccentric shaft, but I had to use needle nose pliers to gently pull it out.

Next I bent the thermal pellet as much as the video and wasnt able to push it all the way into position by hand, but had to thread the eccentric shaft bolt in to push it into position.

Then I heated it up to test the expansion inside the eccentric shaft and it worked because that wax thermal expansion in sealed cylinder exerts a lot of pressure.
However, the pellet was now stuck in there when I removed the eccentric shaft bolt and had to be removed with pliers and force revealing scoring on the side of the pellet.

So, to recap.
If bent and installed the thermal pellet would seize shut (full oil flow to rotors) not open as the sealed thermal expansion of wax is a far stronger force than the return spring.

If bent that much the pellet would be stuck in the eccentric shaft as mine was when the front bolt is removed and not being pushed all the way out of the eccentric shaft by the return spring as seen in the Vargas video.

Further; as I observed, the thermal pellet would have scoring on the OD of the pellet where it rubbed the inside of the eccentric shaft from just 1 expansion event and the pellet in the Vargas video is still perfectly smooth.

My conclusion is after the cut scene with the thermal pellet shown still sticking out of the eccentric shaft the thermal pellet was remived and bent and then filming resumed.

Much like with the off camera flywheel removal and partial reinstall (not fully torqued/ re-seated on the taper) before the end float check.
_______

The claims of metal in the oil and oil pump scoring.

They claim there is metal in the engine that scored the oil pump. They claim it is bearing material from failed thermal pellet operation.

Every single cold start of a rotary is done with the thermal pellet bypassing oil and it doesnt trash the bearings. Why would Mazda's combustion check trash the bearings even IF the thermal pellet could be stuck open in bypass mode?

The silver bearing material is called babbit and it is a VERY soft sacrificial metal that overlays the harder (soft) Copper bearing.

It is not very unusual to have babbit in the oil of a fresh factory spec rotary engine rebuild. Very tight bearing clearances are used on stock engines and the babbit often gets pushed out of perfectly round as the bearing is pressed into the stationary gear and ESPECIALLY the rotor.

This is why Mazda recommends a longer break-in for a rebuild when using new bearings and recommends against the use of both new bearings and stock bearing clearances in a performance rebuild.

Babbit metal is so soft it harmlessly travels through the oil pump and engine.

If there was another type of metal in the oil then it would show there is an issue with their own diagnosis in the video.

Another metal in the oil would also indicate Mazda made a mistake on assembly, but because of what I see as an overall shadyness of the Vargas video and their motives I am inclined to not believe their "findings".

Funny you should mention that regarding end float.
When measuring end float on my new crate engine I also got 0,10 mm end play or 0,0039".
I did disassemble mine and I ended up replacing 7 side seals due to being over the recommended limit. It was also nice to clean up all the coolant passages and do some port matching while at it.
I also replaced the spacer, which was marked "C" with an "A" spacer and got 0,06 mm / 0,0024" end float when complete, which is just inside the '94 engine manual spec.
Would be interesting to see if anyone have access to the 99+ manual to confirm which spec. is correct.

Did anyone notice that it appears the TPS sensor is on the list of remade/available parts? N3Y313SL0? N3Y113SL0?

yep, they brought that back in like 2019 or something.

if anyone would have the current spec, it would be Mazda... its actually possible that the endplay changed when the Rx8 was introduced. the rotors were changed in 2003.

honestly though..... considering their process, if theyre setting the end float to that spec then its more than likely thats the correct spec. i dont see it being possible that they have this number wrong and then 2 separate people saying thats the spec their motor has. this is very deliberate. you can see too they have that giant wall with the assembly reference material on it so everything they need to do is printed.

I guess this is why we can't get 20B and 26B crate engines.

Hard to believe Rx8s have been out of production for 10 years already, still think of them as only stopping a few years back.

2 engines per employee/day, not withstanding currency fluctuations, that might explain why a new short block over here has basically triple/quadrupled in price in ~ 20 years.

Suppose it could be worse, old $500 SR20s dried up some time ago, new production ones, I read somewhere the other day, come in at well over 1 million yen, before even thinking of freight and taxes!

Brand new 13B-REW still looks like an attractive option in USA.

New part # N3G1-02-200 for the high power 1999+ spec manual.

$7,000 full retail,
see it for as low as $4,750 online (but I dont know if pricing is current).

If you have Mazda Motorsports account (easier than ever to get) its still less than that lowest price I found online.

Great support from Mazda.

Some company on FB had them locally at Au12.5k the other week.....but it was on a bit of a price dreamer page. With the weakness of the yen recently, would have expected them to get a bit cheaper - but container transport is still messed up to make up for it I suppose.

currently it’s listed as $4900 USD plus shipping from Motorsports

wrt to ordering the engine though, they’re listed as backordered and you’d have to place an order and then wait your turn as opposed to various vendors who did that and have them in hand for a higher price.
.

I tried to get an 13B-REW into my cart as "guest" /not logged in without any luck.

I was able to get a 13B-MSP for RX-8 in the Cart for $4,900 as you say.

I could try calling if I was super curious since Mazda Motorsports staff has always been extremely helpful and helped me find parts.

13B-REW loaded in cart through my account shows back order (so, in Japan or on a boat).

Its ~$700 less than the 13B-MSP.

Very cool, thank you for sharing.

i changed my build biz Jan 2020 so as to only build CPR Spec motors from new blocks. my routine prior to that was always to fill out a 6 page spec sheet on incoming motors. i continued that with all the new motors. consequently, i have detailed data on a bunch of new blocks.

my general takeaway is that they are, with two exceptions.... jewels. i have taken loads of pics of them just to marvel at how beautiful a new motor is. there is a dot of paint on every fastener. when i say every, even the tiny stake for the main bearings. as to bearing clearance, someone mentioned "tight." not on my motors... perfect. rotors typically less than 2 grams apart. endplay approx 2 thou.

the two items i find not to my liking are:

i find it always easy to remove the oilpan. the silicone, presume "Mazdabond", is gummy, sticky, pretty runny. it gets all over my sockets. i can only assume that it sets up with heat as the motor runs... if i remove a pan from one of my builds w Hondabond it is really hard to separate. you can remove all the bolts, and with the motor upside down easily lift it by the pan without it separating. Hondabond and Mazdabond are made by the same company, ThreeBond, located in West Chester, Ohio... clearly to two different recipies. this certainly is not a big deal but it is always a sloppy mess. of course if the new Mazda block pans don't leak it is O K not an issue.

the second issue, again, is not a deal breaker but is of a bit more concern. sideseal gap. on over ten motors it is consistantly all over the place. i find more than 6 of the 12 sideseals to be gapped at 5 thou or more. most motors have one or two over 10 thou.

here is one at 12

https://imagizer.imageshack.com/v2/1...922/ktt4mt.jpg

i find it interesting that sideseal gapping is one of the few processes that can not be automated. everything, in my experience with the new blocks that can be automated is PERFECT.

enter humans and...

a recent Mazda funded MIT study attributes rotary compression as following:

apex seal 65%
corner seal 25%
sideseal 10%

my experience confirms this as motors with totally collapsed corner seal springs are generally down 30%. so the sloppy corner seal gaps, again, aren't deal breakers but i redo all my motors to between 1 and 2 thou.

bottom line from someone who is pretty anal the new motors get a big thumbs up and, as tomsn16 posted, for a variety of reasons make more and more sense from a financial perspective.

the sealant Mazda uses has changed a couple of times. the 8527-77-739 sealant that was used from the dawn of time to 2003 is a Takata SH780M. the current number R2Y1-10-431 or in the USA 0000-77-1217-ES is a Three Bond TB1217D
no experience with the new stuff, i've been using permatex.

if you have the chance Howard go and ask a permatex guy about sealant, its a more interesting conversation than you'd think :)

As Howard mentions about the sealant, mine as well was gummy and not even cured around the dummy motor mount bolts...
I wonder if they might have used a sealant which has passed it's expiration date - YES, sealants can go bad over time and are stamped with a best before date, like your milk..
After much research and trying to get ahold of "Hondabond" here in Norway/Europe I gave up and ordered a tube of the OE Mazda sealant TB1217D.
Got the tube, and to my surprise, it had expired by almost 6 months...
Handed it back in and got a new one, with good date.

After doing my oil pan the sealant is completely hardened and feels "proper" for the lack of better words.

Do you have any insight as to the end float mentioned earlier Howard? As I said, mine were 0,10 mm/0,0039" from factory w/"C" spacer, and after I changed the spacer to an "A" I got it within 93-95 spec (0,06 mm/0,0024"), wondering if I should put the factory spacer back in...

These are probably built better than the original production line ever did. First, there's an updated process. There's no assembly line to keep up with. They know the people buying them are paying close attention to them, so they only have these 3 guys building at the highest quality. Mazda's probably breaking even on them at most; it's more fan service than anything.

^That's an interesting theory, but I doubt an expired "best by" sealant would be an issue in this case. As indicated in the OP's linked article, Mazda's logistics process (i.e., parts supply for the builds) is highly automated and appears to have solid quality assurance & configuration management built into it. So I think the odds of an expired sealant (or any other sub-standard part for that matter) getting into the build process are pretty slim.

I think Howard's theory that the Mazda sealant sets up & hardens after several heat cycles is spot on, and I have one sample of this from personal experience. Just prior to purchasing my FD from its PO, he installed a new Mazda 13B-REW crate motor, and he didn't open or modify it in any way. I test drove the car with maybe 25 miles on the new motor, and then subsequently put perhaps another 100~200 miles on it post-delivery before pulling the motor. I was doing a full rewire of my Link G4+ ECU, and wanted to pull the motor to install a Banzai racing oil pan brace, and port match the intake port runners to the 13B-RE Cosmo LIM the car came with. When I pulled the oil pan, the sealant was hardened & cured all along the perimeter, though pliable enough that it just needed a bit of persuading with a razor cutter to free the pan from the block. Sealant that got into the bolt holes remained pretty soft though, and had to be chased out of the threads prior to replacing the bolts with the studs that come with the pan brace kit.

I was told that by a rep there. At the time I had an account but was absent on results due to a health issue keeping me sidelined. So possibly I misunderstood or that person misspoke because I was an established racer in the program since joining back in 1993 or so. I could not access OEM parts without results though, only the race parts. So you must of replied while I was editing my post. Because I had realized that my situation was not the same thing as no account and deleted a lot of my initial post for that reason. Sorry about that.
.

the rotary engine factory is lucky enough to be in a separate building, so they have a whole building that does nothing but make rotary engines, so you might as well make rotary engines.

the piston engine people are/were less lucky, the Skyaktiv program started with a remodel of the main factory, which is where the piston engines are made. they found that after the castings were cast they spent around a week going to the 14 odd machining stations. after the revamp the engines castings go to two stations. the thing with this is that if you have a B6 or BP engine Mazda, like what is in a Miata, those are NLA, i'm not sure they could make them anymore so we are lucky that the rotary plant is how it was in 2003.....

I have the entire collection and the 7/1994 manual states a Standard Value of 0.040-0.070mm and Limit Value of 0.090mm for End Play in Section TD (Technical Data). The 2/1992 13B-REW Engine manual has the same number in it too. These are the exact same numbers in the FC's 13B Engine manual from 1985 too. None of the later manuals for the FD change these numbers.

This change would imply that something was learned while developing the 13B-MSP for the RX-8. I would guess that due to its high-revving nature, scalloped rotors and the looser 0.10mm End Play numbers were made to account for extended high rpm driving and these changes were then "backported" to new N3G1 engines. As for why, I'd guess as a safety measure, but a more experienced engine builder would be better equipped to explain the reasoning than I can.