Jump to content
Strawberry Orange Banana Lime Leaf Slate Sky Blueberry Grape Watermelon Chocolate Marble
Strawberry Orange Banana Lime Leaf Slate Sky Blueberry Grape Watermelon Chocolate Marble

Mazda 626 Hatch - Labour Of Love

Recommended Posts

By Request !


I have always liked the 626 Hatch and to date have owned five of them, (pics here) three consecutively at the same time.

Since helping the original MrSpeed (Patrick or Adlorin) put together a supercharger kit for the Probe and testing it first in Australia I kept

building on new ideas and changing my goals.


My end result was my 1995 626 Deluxe HatchThe deluxe edition came standard with a few extras hat the standard 626 Hatch here in Australia didn't. 

  • Digital Climate Control
  • Full Leather Interior

My first upgrades consisted on Suspension and braking components as my original goals were always to have a nice looking sleeper so i didn't want the exterior to look much different. I Tried the many different suspension options available for the probe but was never happy with the results. Instead I finally decided I liked the custom feel of the BC Coilover kits with a matching spring combination close to the later released Tein kits.  of course knowing i was going boosted i also needed to make sure i could stop safely. This was never going to be a track car but my daily driver, so i didn't need to go fully overboard. The Mazda 6 Breaking system was chosen in the end as it was almost a direct bolt on

but more importantly it looked stock (licensing restrictions can be a biatch sometimes).





If you have a good look you will notice one other big advantage with the coilovers. No nasty perch to rub against the tyres/rims.

hence you can run a wider profile without interference.




The next route was of course some nice rubber that matched the profile of the vehicle and worked well with the suspension. I trialled several 18" setups and for a few years utilised the low profile Pirelli 225/40/ZR18's on a set of 18" Advanti Rims. These looked good and felt good and made sure the car handled like it was on rails. I encountered one minor problem though. every time I went on a cruise and run across poor roads with many potholes or rough surfaces I was continually scratching up the rims (not to mention you felt every bump) and having to spend $$$ fixing the damage.


Eventually I downgraded in size to a 17" Mazda 6 Rim running 235/45/17, this is what I'm still running today and gives a nice stance as well as excellent performance yet doesn't scrub out every time I hit a pothole (no sore butt cheeks now).



(here it is after some more panel work in the shop)



The original motor was of course a KLZE auto with custom exhaust , (ebay headers with a custom catback running 2.5" pipes with redback CAT and muffler). but the auto just had to go. That was the next conversion.


I shopped around for another donor vehicle with an MTX transmission and upon finding it pulled the necessary bits to convert this to manual. The only problem I encountered was the wiring loom. Being the deluxe edition it had far more wiring than the stock vehicles to utilise the digital climate control.  So I had to use a 93 loom and dismantle the stock loom to match pin for pin what was needed without the ATX extras. Now that was fun (sarcasm).


This then became my daily driver for a while until my other plans came to fruition. A KLG4 Hybrid supercharged engine. 

But until then I simply stuck with a SC KLZE using an Eaton M62 in the AC location.



Fabrication stages of the AC Bracket



The same process but showing the first overdrive Crank Pulley that was CNC'd by MR Speed (Patrick), again a lot of assistance and co-operation from many members

went into the fruition of these projects.



A comparison shot of the OEM 36-1 timing pulley next to the overdrive pulley. This increased

the Eaton's Boost from 4.5psi to 8psi without the IC



Around the same time (2006) a member on Probetalk posted his findings on a new set of custom rods that could increase the longevity of a boosted KL engine. the specs listed below

  • H-Beam style rods; pretty much the industry standard for cost effective high performance applications
  • Machined from 4340 chromoly steel forgings
  • Piston pin bushing left undersized for hand pin fitting
  • Finish ground on Sunnen cross-grinders
  • Millenia-S size piston pin bore
  • Millenia-S/KL size crankshaft end bore
  • Millenia-S center to center length
  • Bolts are ARP 8740 series with 12 point heads and have been increased from 8mm diameter to 7/16
  • Like 1

Share this post

Link to post
Share on other sites

Then I Started the block modifications in preparation for rebuild and assembly. This was done by a local machine shop that gave me full access.



Boring of the block to 20 thou oversize.



Notching of the block in preparation for the rods.



And the pic on the right is the final balancing of the crank with pistons. rod weights being calculated.



Crankshaft machining  after the initial balance process was calculated. and to the right the beginning of the lube and assembly process




Some quick snaps of the preparation and assembly. The Notching you see in the pic is to allow the rods enough travel from the bottom end up to allow the piston/gudgion pin room to secure the piston

before it can be lowered back into the block. the big ends of the rods will not fit down the bores no matter how hard you try.


A youtube photo album compilation can be viewed Here

Share this post

Link to post
Share on other sites

That is one beautiful 626.  Glass looks brand new.  I want to hear about the KLG4 supercharger.  You put a KLG4 in a 1995?  I want to read all about it!  I know you have a link to another sites that details all of it but I think it would be better to post it all up here so everyone can see and discuss it.  You've done more to your car than most dream of (including me).  You are an amazing enthusiast that has done some amazing things.  I know you've shared all of this already with the Probe community but it's not really a Probe is it?  It's a 626 and deserves for everything you've done to be archived here!  Unlike other sites any pictures you post here are saved on the server.  You'll never have to worry about going back to a topic 10 years from now and all of the pictures are missing.  Not going to happen on this site!  Archive it for all time properly here!

Share this post

Link to post
Share on other sites

img-383090-1-c4f9c79294d15b69ebf39c61f82 img-383090-2-19b0179314c9cc0242c2312dab5



From the above picture you can fully understand the need for the notching in the block and how the assembly process must be followed.

Other mods that were made to the block and components in preparation for the build is the oil gallery re-bored on the block and the crank as well.

to this bottom end rebuild the only other change was shimming the oil pump to increase the pressure relief and hence increase the oil flow volume.


The next stage in the build was to consider the heads and the route to follow to help achieve the gains I wanted. For this the original choice was a KLZE set of heads

port matched to a standard IM and changing the valve train to more reliable lightweight components but retain the KLZE/KL31 Cam profiles.






The heads were basically stripped and the valve train components replaced with KLG4 springs and retainers. as well as the valves, and crunch space being highly polished.

The only other mod chosen to reduce overall rotating mass was a set of adjustable cam pulleys. (be aware that the adjustable cam pulleys were not chosen for the adjustment capabilities)






Knowing that the engine was going to be boosted and another weak point might be the stretch to yield head bolts myself and a couple of other members that had purchased the rod/piston combinations looked into the possibility of some ARP head studs. At the time ARP Fasteners were only too happy to do three custom orders but this wasn't good enough for myself. Honda guys had readily available stud kits available for their vehicles why couldn't we.


So In mid 2008 I contacted Chris Brown (Director of Specialty Products) at ARP Fasteners in Santa Paula, CA to investigate what was required to make this possible. Luckily with the thanks to Probetalk and other forums many KL enthusiasts had been in contact for custom orders in the past and they deemed there was enough interest to get the ball rolling so to speak. 


The first requirement from ARP was a KL series engine that they could utilise for R & D purposes. A donor engine was obtained and the process begun. This was basically a long process so rather than post the email updates I received on a monthly basis from Chris I'll just say that nine months after the process started the parts were produced in bulk kits and supplied to resellers with a full off the shelf part number. ( A big thank you to crossoverauto for being one of the first to stock these for us)



Share this post

Link to post
Share on other sites


Pictured above assembling the valve train, lifters and cams




And that was the first build assembled and pretty with my nice powder coated parts to match.


Having completed this build i started on investigating the electronics aspect for engine management as i figured i could test and prepare everything i needed on the current working

daily driver with the KLZE without risking damage to this build.


The first choice of course was the Megasquirt system as so many others on Probetalk had experience with the product if i needed help or assistance.

So I purchased a couple of MS1 Ver II and then version III kits and even the MS II when it came out.





And installed it into my test vehicle.




and began testing with cro. I new I wanted to go fully synchronous in both spark and fuel eventually so i was already pushing the limitations of the MS1 and that was why i was always

upgrading to the newer versions all the time. I wanted as large as a map as I could build 32x32 ideally with full fuel and spark control. The 36-1 crank from the KLG4 at least gave me the clarity

on trigger points and I just used the Home CID from the stock distributor to give  me the 720 degree firing events required for full spark and full control.

Share this post

Link to post
Share on other sites

But I started with the coilpack from Mazda 626 (KLG4) first for spark control with Millenia S injectors and the Eaton M62 setup first. Following is some more pics of the different testing stages

of this period.



Above is one of the driver circuit boards I made for the coilpack. During the testing of this circuit I burned out a couple of the transistors during road tests. Eventually I traced this back to an intermittant

faulty coil in the coil pack but after the burnouts I revised the circuit to utilise IGN drivers instead as they had built in protection. (More pics to follow on that later)


Share this post

Link to post
Share on other sites

Once I had a working MS base map I started using autotune with the wideband to fine tune more of the map whilst I was driving multiple trips. This worked beautifully but I hated having a complete distributor hanging out and not being used just to have a home trigger on the CAM.


So I began purchasing multiple CAS units from different vehicles to find something that closely resembled what I needed and would fit into the stock distributor location.

Thus was born the modified CAS unit from a Miata 1.8L engine.



The above CAS unit was almost perfect. My machining the outside diameter of the cylindrical part until it met the base of the sensor block it gave enough room

for engagement on the CAM and gave me both trigger events for the engine management without the ugly dizzie cap and leads.



As shown above. I simply removed the trigger assemblies from the oem distributor to install into this CAS unit. no issues.




And here is how it sits on the heads on test fitments.




What you cannot see is how it is secured. under the cam covers there is a half shell that is mounted on that end where the distributor normally bolts in. For me I simply rotated the CAS unit until My home signal was where i needed it to be and drilled a countersunk philips screw through this cap and into the CAS unit. The timing never moved after that :).

the more diligent of you may also have see from some of these pics that I had already decided to go an alternative route for the ignition. More on that shortly.

Share this post

Link to post
Share on other sites

The next stage I decided to do was continue on with the ignition aspect. Obviously I had to rebuild my driver circuit to drive six simultaneous coils to go full synchronous but I also decided that

Coil On Plug Would be a nice option as well. So after trialling several COP units I decided on a set of CBR coils and wiring loom from xtremethings.com


img-383194-1-c8dbdb2f413cc80c64c7b6b6b96  img-383194-2-60e29411f35ba38b1cebb11f4af


Here you can see the COP units and the wiring loom, the only drawback to these were I had to use Marine Plugs as there called. You know how some spark plugs you can screw the 

tip off where the high tension lead normally attaches. then this is what you need.

Share this post

Link to post
Share on other sites

Once again whilst rebuilding My goals changed again. The biggest concern was the KLZE heads. because the heads are fitted with two different size lifters (HLA) a 30mm and a 33mm and the difficulty I came across whilst trying to source the 33mm lifters I knew then that any future rebuild would create this same problem, not to mention the expense. So after a few drinks and a late night in phone calls back and fourth to US and Japanese Suppliers of Lifters I decided my best course was to utilise the more common 30mm lifters. But to do this I had to go back to a standard head...


After further consideration I decided to scrap my CAS setup and go for a KLG4 Head and IM and then I could use the rear cam sensor, not have to port the runners to match and I could save myself some time.

So I purchased a couple of complete top ends (again thanks to Michael aka KLZEPORSCHE) for his assistance in obtaining and shipping some of these items for me.


Let the games begin.


img-383438-1-63ef143b1135f146234436a4aef  img-383438-2-fc3971e6e26559588c3b93cc970

EGR hmm whats that for ?


img-383438-3-b323e6c38c218bbbbca5a2818fa  img-383438-4-f373d3d7860df2d93937c5ff346

A bloody ugly waste of time thats what it is!


33c1c58a8576b628ef0e995d8057b7fdc8eeec47  698341219cefd326ec951d9afde14e316625a0cd


Lets start cleaning it up... damn this is time consuming.


04c5d89a03c35ae8e844ff23112aeaa2c686ba4d  2e3abd7753b2d0e52da3fca194f9a63e38398f58


8f65494f3fa24713fc7d054d261f98cff66c0894  500bfeb741519ffb0a5a4f58123072e246e89b94

Vris, who needs that on a boosted engine. Hell we don't need all these mount points in the casting either.

Share this post

Link to post
Share on other sites

And it seems my folder for the IM is corrupt so All i can give you is a few pics of it complete with the ceramic coating.


img-383441-1-cbd898727d3abd74888af2dfa75  img-383441-2-227732042ddb09ba672c7e7696d


Did I mention the ceramic coating... All intake and exhaust components were ceramic coated inside and out to attempt possible heat buildup.


The new heads were prepared the same way as the klze heads with one minor exception. I purchased a set of Colt Tri-Flow CAM's for the build as well.

Share this post

Link to post
Share on other sites

Having Already Purchased what I thought would be my last blower I'd ever need (Rotrex C30-94) I began the mockups for the Intake and exhaust pipe work.

The ebay headers were nice but had limitations on the secondaries as well as the problem of turning the engine bay into a heat soak. My first though was a wrap solution but it looked ugly as

and i was concerned about it trapping moisture. My final solution was to ceramic coat everything first to see if this would work.


0ed4e78f969f8e1792749488d042052c6afe6916  1def92c5c16261cbc37e456648ff67ac960e8060


Tie up and spot welds in place


65bbb497a301d5bfa9225153edff0686ba52e28e   c9e716455c15dc8ec38591ef2d763eadcfe77d92


Almost finished.


88154e2975e2dfa00dd78395042c3195794b3144  204333e47e542259dbc02b8b12853532e3936298


Welding done. Now onto the ceramic coating.


7541937976d39221ec6e7bf9c2d18b1ba3045d6d  d0f2bd2ec74309574550c4a39263ca4fc9d6c8ca




Purty ain't it ?

Share this post

Link to post
Share on other sites

Then onto the IC and discharge piping. Again more creativity here for mounting both the IC and relocating a power steering cooler as well.


img-383576-1-bc4ed3e0de1595fc578120ba4d2  img-383576-2-7f77d86e4eb63c0bb999f47f8f4


Front shot showing position of IC


e8a1cee804874b6fb330db5b383ccc494b16202a  d391ecfb96968ac3baea47dcd93ca308d1ee2316


Showing clearance behind the bar after modifications to the front radiator support. (yes it will survive a front on collision without the front end collapsing)



Power Steering Cooler in new mock up position, in some of the photos above you can see it under the IC but this left me no room for the blowers oil cooler.



38451037941d6e6ddc06e3e078d6fabab0ef61bd  cc5f2cfd48ece103d3bccdf9f4f29eac82c5f569


Front Stance with it installed.

Share this post

Link to post
Share on other sites

One important aspect I forgot to mention earlier. Obviously with boost you must be able to supply the fuel.

All my vehicles I automatically upgrade the fuel pumps to Walbro 255's with new regulators. But of course you need injectors to be upgraded to supply as well.


For my first build I opted to use Mazda RX7 Primary's at 550cc. at the increased fuel pressure of 43.5 psi you get a bit more.

00661a831867fbd3bdb3d9c4b4a92992f4a521a3  33fa0f0755c709448b8ef7c9aaa0b907914eebcd


564 CC not bad.


33fa0f0755c709448b8ef7c9aaa0b907914eebcd  bd034aaefb4f6ac931fc9790e683b4dd48fb324d




Again going with the Royal theme (purple and Gold)

Share this post

Link to post
Share on other sites

And the Discharge Piping was basically the same process as the exhaust.


img-383579-1-df237c91a4583dc489fbc42b0db  img-383579-2-8a50fb83260d0f02aa7032f390c

From the blower to the IC and from the IC to the IM with the BPV flange in between.


img-383579-3-1db56705cc29a6878d8272cebc9  26daf36176eff5e73d5d1c6dada84bd58a221bab


And then.


d41dd81f88e0550918786e610ed270fcece8d061  6ed88db68a4e61816c280995b96181dbde8b93a9


Then Ceramic coating and re-installation

Share this post

Link to post
Share on other sites

After putting down 5k km's on the engine I noticed that I was having issues with heat and large amounts of blowby in the crankcase ventilation system.

We have no definite answers at this time but I discovered the rings supplied at build time were not what I requested. The Engine shop simply installed them anyway and forgot to mention

it to me.


but basically the ring gaps are larger than they should be for a boosted engine and they are standard "high performance" rings not the stainless file to fit I had requested.


At final destruct time the engine was running a Haltech Sport 2000 Engine management system paired with a M&W 16B CDI unit to fire the COP's in full sequential mode.

The M&W CDI unit will shortly be replaced with a newer single wire unit but until the rebuild is completed it will be standing anyway.


The Clutch is a SPEC stage 5 (yes i know , Overkill) but with the amount of $$$ spent it will probably remain so.


Due to financial constraints this build/rebuild is now on hold. Im on a pension (forced) and had to invest the last $$$ i had in getting a daily driver organised for the transport of the wife and kids.

This at least is now finished (except for a rebuild transmission) but is usable until such time as I can get my wife the care she needs and I can return to work.




Current daily driver

Share this post

Link to post
Share on other sites

That's a pretty epic build and documented very well.  Thank you for all of your efforts both on the build and documenting.  So let the questions begin.  Why cut up the manifold EGR ports and how did you close them.  It doesn't show and doesn't say how you closed them.  Leaving them open obviously would introduce massive vacuum leaks so they had to be closed somehow.  Welding?

Share this post

Link to post
Share on other sites

That's a pretty epic build and documented very well.  Thank you for all of your efforts both on the build and documenting.  So let the questions begin.  Why cut up the manifold EGR ports and how did you close them.  It doesn't show and doesn't say how you closed them.  Leaving them open obviously would introduce massive vacuum leaks so they had to be closed somehow.  Welding?


EGR is just a part of the anti-smog legislation that manufacturers had to comply to. Most people can tell you they have there vehicles running without the EGR and still pass emission tests without issues.

For my part there was no way I wanted boosted gas from the plenum leaking past the EGR valve into the exhaust and definitely no way I wanted hot exhaust being recirculated into the cooler charged air after the air intake sensor.


This would throw off the fuel calculations for the engine management as well as waste the effort into ceramic coating the intake manifold to reduce heat buildup/soak.

So this was removed from the IM as were the VRIS butterflies and mount castings. Holes were welded and re-sanded before ceramic coating as part of the preparation process.




this was my first attempt at porting a G4 Manifold. The one you see on the engine now is the second attempt. The second attempt also has the additional castings, mounting points and vris as well as all casting lines machined off

Share this post

Link to post
Share on other sites

The ARP R & D story is amazing.  Thank you and other Probe community members for making a lot of these products available to us.  Those must have been exciting times.  Might want to detail more about the difference an ARP stud has over the stock head bolt.  Would make for a more informative read.


The Miata CAS unit you machined is just a thing of beauty.  Looks so neat and clean.  You mention full synchronous spark for the COP setup.  Are you saying it's been changed to a waste spark system?  I didn't even know extremethings sold COP and ignition systems what the hell.  Did it come with the orange loom or did you add that.  It's a nice clean look.    Ohhhh man you scrapped all that CAS work just to switch to the G4 head and IM.  That's a shame, it would have been beautiful.  I don't see a distributor or coil in the finished images though.  So what is driving the spark with the G4?  Looks like the wiring is going down next to the fuel filter.  I realize the MS can drive the ignition signal but what's producing the spark.... oh wait it's COP holy shit I'm an idiot.  That's so clean and genius.  COP and MS is soooo clean. :thumbsup:   So why all the work with the CAS was that still on the stock ECU and ignition wires at the time?  OMG so many questions and goodies.  Your build is so neat and clean it's obvious that you are highly proficient with this stuff.  I want to get into MS someday looks like a lot of fun.

Share this post

Link to post
Share on other sites

ARP head studs or any well made Stud has several advantages over stock head bolts but mostly are used in high HP engines.

Standard head bolts are a stretch to yield design. The more you attempt to tension them the more of the force has to be applied into the torsion of the entire bolt, risking snapping and fatigue

after multiple uses. A stud on the other hand is applying the torque directly to the clamping force of the materials. So in the case of a head you are able to use more clamping force with less metal fatigue

against the mating surface of the block.

This also can reduce the likelihood of the head warping under possible hot spots and allow the engine to handle far more HP before blowing the heads.

Summit Video explanation


A quick blurb with more detail found in a local magazine write up






factory Tech Representatives are often asked which is better, cylinder head studs or bolts. The answer, invariably,

depends on the installation. On many street-driven vehicles, where master cylinders and other items protrude into the engine

compartment, it's probably necessary to use head bolts so that the cylinder heads can be removed with the engine in the car.

For most applications, however, studs are recommended. And for good reason. Using studs will make it much easier to

assemble an engine (especially a racing powerplant which must be serviced frequently and quickly!) with the cylinder head and

gasket assured of proper alignment. Studs also provide more accurate and consistent torque

loading. Here's why. When you use bolts to secure the head, the fastener is actually being "twisted" while it's being torqued

to the proper reading. Accordingly, the bolt is reacting to two different forces simultaneously. A stud should be installed in a

"relaxed" mode- never crank it in tightly using a jammed nut. If everything is right, the stud should be installed finger tight.

Then, when applying torque to the nut, the stud will stretch only on the vertical axis. Remember, an undercut shorter stud

will have a rate similar to a longer, standard shank stud. This provides a more even clamping force on the head. Because the

head gasket will compress upon initial torquing, make sure studs and bolts are re-torqued after me engine has been run.





The original design scope was with the MS. With any four stroke engine you have 720 degrees of rotation for a complete cycle. If you are only firing from a crank trigger you have no way of knowing if the engine is on an intake or an exhaust cycle, so with only a crank trigger you can only do a wasted spark setup. However if you utilise a Crank and a Cam signal you now have in a sense a home signal which can be used to indicate your on a compression stroke. Hence you can now do full synchronous timing events and no longer need wasted spark setups.


The MS was eventually scrapped as well and the whole system updated as mentioned in the last thread. I wanted something with more support and a clear upgrade path for future mods. Haltech offer easy enhancements

and work closely with M&W Ignition systems. Ask any race car driver these days if they know what a MS unit is, then ask them if they know M&W CDI systems. Most likely they are more in use than MSD ignition systems

on most new racepak's.


As soon as I went down the G4 head path I automatically gained my cam trigger for the home sense circuit, so the Miata CAS was superfluous. plus id looked neater without it.


The loom on the COP's was also custom made by Mike for me using high grade teflon coated wiring. with double shielding so no ignition spikes or radio interference. 

img-383642-1-c8dbdb2f413cc80c64c7b6b6b96  img-383642-2-60e29411f35ba38b1cebb11f4af


A few more pics.

Share this post

Link to post
Share on other sites

Ah I forgot you mentioned the cam sensor on the G4.  That's why you switched.  G4 heads on the same DE block no problems there?

Share this post

Link to post
Share on other sites

Ah I forgot you mentioned the cam sensor on the G4.  That's why you switched.  G4 heads on the same DE block no problems there?


no, exact same block. only difference normally would have been the compression ratio but as my build was custom I already had the 9.7:1 so easy as.


And for you reference here is a vid of the early testing of a wasted spark setup

Share this post

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...