1975 Mazda REPU - Build and Repair thread

chuyler1

goes to eleven
Moderator
V
2013 CX-9
Started the truck up for the first time yesterday. It still needs to be tuned but Lui got it to idle. There were a few things to iron out over the last few days but it's all sorted now.

Fuel pump was junk. I diagnosed everything else and had it working at one point but as soon as I upped the fuel pressure it conked out. I picked up an OEM replacement for an '89 Ford F-150 at the parts store that comes with a lifetime warranty.

Alternator wires were reversed so the dash lights weren't coming on. Easy fix.

Throttle position sensor was on backwards. It was symmetrical to mount on either side and the instructions weren't all that clear...but it was another quick fix once we saw the values reversed on the ECU.

Thermostat wasn't opening so I replaced that to be safe.



 

bazooka joe

No words
Moderator
Contributor
V
2016 Miata Club, 2014 CX5 Touring, Honda 1976 XL 175
Started the truck up for the first time yesterday. It still needs to be tuned but Lui got it to idle. There were a few things to iron out over the last few days but it's all sorted now.

Fuel pump was junk. I diagnosed everything else and had it working at one point but as soon as I upped the fuel pressure it conked out. I picked up an OEM replacement for an '89 Ford F-150 at the parts store that comes with a lifetime warranty.

Alternator wires were reversed so the dash lights weren't coming on. Easy fix.

Throttle position sensor was on backwards. It was symmetrical to mount on either side and the instructions weren't all that clear...but it was another quick fix once we saw the values reversed on the ECU.

Thermostat wasn't opening so I replaced that to be safe.



Nice work Chris! Looks great too!


Sent from my iPhone using Tapatalk
 

CTt3P5

New Family Truckster
Contributor
V
2008 MX5 SE 33Y/2016.5 CX-5 GT 42M/2010 3i T 38L/2008 3s GT 34N
I agree on both counts.

Easy fixes are always welcome. What else do you have planned?


Sent from my SM-G955U using Tapatalk
 

chuyler1

goes to eleven
Moderator
V
2013 CX-9
Well once it's running I plan to just start driving it until next winter. No more projects that will take it off the road. However I do have some things I need to tweak...

The fuel lines and fuel filter under the bed of the truck need to be upgraded. I have a filter, I have lines, I just need various fittings and some help taking the bed off.

I have an extra wire going to the front of the car. Either it's going to power the windshield washer, or serve as a low coolant warning. Can't decide if I want the washer reservoir in the engine bay. The low coolant sensor was never connected so...it's no biggie.

I need to work out an oil catch can of some sort, or hook up a PCV valve to the new intake.

Getting the OMP working again would be nice, but I'll just premix for now.

I want to rebuild the wiper mechanism, it's binding and really slow despite now having a full 14v to operate (old alternator was pretty weak).

The audio system could use some attention. Not a lot of room for kick panel speakers but I've seen others put speakers in the doors. I have the seat thumpers and headrest speakers connected but they are poor quality. You know me, I have standards...and I'll need something that can drown out the new intake sound.

Depending on how much power this makes a future project will be to street port the motor and install a LSD. Then, if it's too much power for the suspension I'll consider putting the coilovers back in so I can maybe autox it a few times.

So anyway, plenty of fun stuff to do...but first is the tune so I can drive it. Lui is coming next week. If it goes well, I'll have it at the Mt Washington meet.
 

bazooka joe

No words
Moderator
Contributor
V
2016 Miata Club, 2014 CX5 Touring, Honda 1976 XL 175
Well once it's running I plan to just start driving it until next winter. No more projects that will take it off the road. However I do have some things I need to tweak...

The fuel lines and fuel filter under the bed of the truck need to be upgraded. I have a filter, I have lines, I just need various fittings and some help taking the bed off.

I have an extra wire going to the front of the car. Either it's going to power the windshield washer, or serve as a low coolant warning. Can't decide if I want the washer reservoir in the engine bay. The low coolant sensor was never connected so...it's no biggie.

I need to work out an oil catch can of some sort, or hook up a PCV valve to the new intake.

Getting the OMP working again would be nice, but I'll just premix for now.

I want to rebuild the wiper mechanism, it's binding and really slow despite now having a full 14v to operate (old alternator was pretty weak).

The audio system could use some attention. Not a lot of room for kick panel speakers but I've seen others put speakers in the doors. I have the seat thumpers and headrest speakers connected but they are poor quality. You know me, I have standards...and I'll need something that can drown out the new intake sound.

Depending on how much power this makes a future project will be to street port the motor and install a LSD. Then, if it's too much power for the suspension I'll consider putting the coilovers back in so I can maybe autox it a few times.

So anyway, plenty of fun stuff to do...but first is the tune so I can drive it. Lui is coming next week. If it goes well, I'll have it at the Mt Washington meet.
Lmk when you want to lift the bed off and I'll come up


Sent from my iPhone using Tapatalk
 

chuyler1

goes to eleven
Moderator
V
2013 CX-9
Lmk when you want to lift the bed off and I'll come up
Sure thing, maybe a fall meet or something.

Another update. Let's see...
The base tune Lui put on it last week fouled the plugs. Of course we spent several hours scratching our heads as to why it wouldn't start and actually found a few other issues, none of which got the truck running again...until I went to the store and got some RX8 plugs and a timing light to check the timing.

I've been learning quite a bit about tuning and after messing with a basic MAP (manifold air pressure) tune for an RX8 I realized it just wasn't going to work with the ITBs since there is literally no vacuum once the throttles open. Lui hinted at doing a TPS (throttle position sensor) tune instead and after a few hours of searching the internet I finally figured out what he meant and how to set up the ECU.

A few more hours later the truck idles at 14.7 AFR but then stalls as soon as I crack the throttle. It took some more time to figure out I could adjust the step increments on the Y-axis of the fuel map. Instead of throttle positions of 0%, 20%, 40%, etc, I was able to change them to 0%, 1%, 5%, 10%, 20%, 40%, etc so that as soon as I touch the throttle it adds the needed fuel.

So now I've got the truck driving and I made a lap around the block for the first time this afternoon. I turned on the data logged and I have a lot of fine tuning to do. Plenty of backfires and stalls. But once the throttle is opened, it sounds amazing and pulls much harder.

Brakes are another story. With barely any vacuum, the brakes have no boost. I'll have to take care of that later, add it to the list.

Still hoping to get it to the Mt Washington climb a week from today. Anyone else going?
 
V
'86 RX-7 Base, '79 RX-7 GS
Do you think a larger brake vacuum booster would help? Unfortunately it doesn't look like you have a lot of room in there.
 

chuyler1

goes to eleven
Moderator
V
2013 CX-9
There is more room than the photos may lead to. I'll address the vacuum issue below....

So, the lack of instructions (i.e., no instructions) for the intake, and my relatively beginner status when it comes to these things rears its head again. I took the vacuum barbs that came with the kit and screwed them into the holes they fit into. It turns out those holes are for idle adjustment. Packed away in all the other extra nuts and bolts the kit came with for various throttle linkage options were some idle adjustment needle screws which have the same pitch as the vacuum barbs.

So I took the barbs out, put the idle screws in, messed with adjustments between the screws and cracking the throttle a little and it behaves better now. Ended up with about 2.5 turns of the screws and a slight crack of the throttle. I got it down to an unstable idle at 800Rpm but upped it to 1200 to keep it from stalling.

So that leaves the issue of vacuum. I had side port placed on one of the runners that I was going to use for the brakes (before i found the vacuum manifold) and I grabbed a T to hook both the MAP sensor and brake booster to it. I'm getting a better vacuum signal but since it's only attached to one rotor, the signal isn't reliable. Once I'm driving it's OK, but the ECU is trying to adjust idle and it goes a little bonkers, hence me increasing the idle air for a faster idle. I'm going to tap each of the runners for new vacuum ports but I'm going to proceed with tuning so I can drive it a little. I'm not ready to take the whole thing apart just yet.

Oh, I'll also add the brakes feel better. They were originally only getting vacuum from the rear rotor secondary runner so it's no different than stock. I'm hoping to get a little more brake boost when I tap new ports.

Back to the driving and tuning. I'm learning more each time I go out. The truck is behaving better at back road cruising speeds. I haven't ventured onto the highway yet. One thing that was bugging me on my drive last night was what I call an "accelerator pump" issue. On a carb, when you punch the throttle, it squirts some extra fuel. The ECU wasn't doing this for me and the truck would go lean when I pushed the throttle down, backfire, the continue on just fine. I read up last night on "transient throttle" and this is a know issue with ITBs because the MAP sensor won't react fast enough. There is a separate table to map predictive MAP sensor values to throttle positions. If the throttle position changes quickly (how quickly and how much is configurable) then the system will ignore the MAP sensor for a short period and read this predictive table instead. That table was way off based on the logging I've been doing so I updated it late last night and I'm going to try it out this evening.

If that goes well, and I get a few baseline readings for the fuel map, I can turn on the closed loop mode and let it do some "adaptive tuning" for me.
 

chuyler1

goes to eleven
Moderator
V
2013 CX-9
Here's a quick example from the log. Once I'm at about 25% throttle at 2,000 RPMs the car settles into an AFR between 13 and 14, but getting there it leans out and stalls. You can also see here the erratic MAP sensor reading. It goes from 25Kpa straight to almost 100Kpa (no vacuum) with the slightest touch of the throttle. Normally you'd see that value ramp up relative to the throttle position.

2017-08-01_1556.png
 

chuyler1

goes to eleven
Moderator
V
2013 CX-9
Oh, and there's other stuff I've been messing with this week:

1) The left rear turn signal isn't working. I traced it back to the Hopkins tail light converter I used to combine the signals from marker lights, brake lights, left and right turn signals. It replaces the old set of relays the truck had. I know the converter was working when I installed it because I took video of the flashers and blinkers working. It was a relief to see them work after months of wiring...the only relief I have now is that I installed it with quick release connectors so I can just pull out that section of harness and swap in a new one. I'm going with Curt this time...and I'm going to check it again in a few weeks to make sure it hasn't failed.

2) The tach wasn't working. Turns out I specified the wrong auxilary output so now that works great, including the shift buzzer. In the future, I may hook the shift buzzer to a separate ECU output so I can trigger it at a different RPM. It goes off pretty early.

3) The speedometer makes a ticking sound and flutters. This has been an ongoing issue (other owners report the same). When I rewired the dash I took it apart, cleaned, and oiled it. I'm going to pull the speedometer cable and make sure that is properly lubed as well. It sounds/looks like it is binding or something.

4) I'm not sure I trust the reading from the water temp sensor. I think it reads high because the fan basically turns on and stays on, even after I've been cruising for a while. The factory temp gauge reads fine so I know the truck isn't over heating. I've already adjusted the sensor values based on ones I found online. It only reads maybe 5-10 degrees warmer than my ambient temp when it's been sitting. I may just bump the turn-on temp by 10 degrees and watch the factory gauge to see what happens.
 
V
'86 RX-7 Base, '79 RX-7 GS
There is more room than the photos may lead to. I'll address the vacuum issue below....

So, the lack of instructions (i.e., no instructions) for the intake, and my relatively beginner status when it comes to these things rears its head again. I took the vacuum barbs that came with the kit and screwed them into the holes they fit into. It turns out those holes are for idle adjustment. Packed away in all the other extra nuts and bolts the kit came with for various throttle linkage options were some idle adjustment needle screws which have the same pitch as the vacuum barbs.

So I took the barbs out, put the idle screws in, messed with adjustments between the screws and cracking the throttle a little and it behaves better now. Ended up with about 2.5 turns of the screws and a slight crack of the throttle. I got it down to an unstable idle at 800Rpm but upped it to 1200 to keep it from stalling.
Do you have any pictures of these idle screws? I'm curious since they only method of idle control I have is to crack the butterflies. I'm guessing the screws act like a sort of controlled vacuum leak?

So that leaves the issue of vacuum. I had side port placed on one of the runners that I was going to use for the brakes (before i found the vacuum manifold) and I grabbed a T to hook both the MAP sensor and brake booster to it. I'm getting a better vacuum signal but since it's only attached to one rotor, the signal isn't reliable. Once I'm driving it's OK, but the ECU is trying to adjust idle and it goes a little bonkers, hence me increasing the idle air for a faster idle. I'm going to tap each of the runners for new vacuum ports but I'm going to proceed with tuning so I can drive it a little. I'm not ready to take the whole thing apart just yet.
I was told to avoid MAP tuning and use alpha-n (TPS) only. The internal MAP in my ECU is strictly being used for barometric correction.

Oh, I'll also add the brakes feel better. They were originally only getting vacuum from the rear rotor secondary runner so it's no different than stock. I'm hoping to get a little more brake boost when I tap new ports.
Do you have a check valve on the brake booster vacuum signal? I couldn't see anything in the picture that looked like one. I'd definitely tap each runner for a vacuum signal. On my setup both of my runners are tapped and feed into a vacuum block that supplies the brake booster. I also have two other ports for MAP and the fuel pressure regulator should I decide to use them.

Back to the driving and tuning. I'm learning more each time I go out. The truck is behaving better at back road cruising speeds. I haven't ventured onto the highway yet. One thing that was bugging me on my drive last night was what I call an "accelerator pump" issue. On a carb, when you punch the throttle, it squirts some extra fuel. The ECU wasn't doing this for me and the truck would go lean when I pushed the throttle down, backfire, the continue on just fine. I read up last night on "transient throttle" and this is a know issue with ITBs because the MAP sensor won't react fast enough. There is a separate table to map predictive MAP sensor values to throttle positions. If the throttle position changes quickly (how quickly and how much is configurable) then the system will ignore the MAP sensor for a short period and read this predictive table instead. That table was way off based on the logging I've been doing so I updated it late last night and I'm going to try it out this evening.
It looks like you're on the right track. Haltech has a "Transient Throttle Enhance" feature that does exactly this. It looks for a certain rate of change in throttle position and adds extra injection cycles to compensate.
 

chuyler1

goes to eleven
Moderator
V
2013 CX-9
On the diagram, the idle ports are right below the fuel injector port. Inside the throttle body there is a passage that routes past the butterflies. After a cold start last night I adjusted them to 2.5 turns each.

https://www.jenvey.co.uk/images/stories/virtuemart/product/TB45_D.pdf

Adaptronic has a lot of idle control features but I think they all rely upon having an idle air control valve. I'll have to decide whether I need to bother with that. If I can cold start with partial throttle I can probably do without it for now. It's a fair weather vehicle anyway.

No check valve on the brake booster. With the brakes and MAP on the same port I'm getting reliable signal that will work well enough for a TPS tune.

After tweaking the transient throttle table I got the motor to rev much better. I am definitely on the richer side of that setting because there's a slight lurch when I first touch the throttle at cruising speeds. I'll spend more time on that in the future but at least it doesn't lean out and stall every time I try to touch the throttle.

I'm definitely going to tap the runners for a combined vacuum. The vacuum manifold I have is perfect for connecting the MAP sensor, FPR, and brake booster to 4 vacuum ports...I just need the right ports :)

So at this point, I'm going to turn on the rapid learning mode on the ECU. I've got a few different zones dialed in on the map and it runs and drives at every speed. Sounds like a beast at full throttle on the highway and pulls to 8,000 no problem. I didn't push it further. Power doesn't really seem extreme compared to before. More noise, but it's not to the point where it'll brake loose the tires. I think I'll need to street port the motor this winter if I want that.
 
V
'86 RX-7 Base, '79 RX-7 GS
Random question: have you noticed any fuel in your vacuum lines? Despite having my vacuum manifold as high as possible, above both the ports on the intake manifold and throttle body, I'm finding fuel in my vacuum lines. I have those little inline air filter things and one of them is pretty saturated after 650 miles of driving. I've verified with my data logs that I'm not getting anything above atmospheric pressure so I'm not sure how it's getting there. One thought is some kind of wave reflection causing reversion based on my manifold design. Either way, it's going to be annoying/expensive to keep replacing these filters every 1000 miles to keep fuel out of my FPR, MAP sensor, and brake booster.
 

chuyler1

goes to eleven
Moderator
V
2013 CX-9
Funny you should ask. I noticed just the other day that there is a bubble of fuel in the clear vacuum line going to my ECU’s MAP sensor. Pretty much all the lines have some amount of fuel moisture but I don’t think it is affecting the vacuum reading.

I forget where I left off with my trials and tribulations so I’ll review what I’ve been messing with lately.

First, I had additional ports put on the intake so I could get a decent vacuum signal from all 4 ports. This works very well and I was able to get a good road tune on the motor, enough to drive it around if I was gentle on the throttle.



Next I spent some time mucking with the idle screws on the individual throttle bodies and the adjustment screw that positions the butterflies. There was a conundrum though. The motor idled best with more air coming from the idle bleed screws...but that made it practically stall when you first touched the throttle. I could only get it to idle at 1,000 RPM. Any lower and it had the propensity to stall when i came to a stop. Then I had a bit of a breakthrough.

Multi-stage injection: The Adaptronic 440d, when used on a rotary can control primary and secondary injectors separately. Although I’m opening all ports at the same time, all 4 injectors don’t all necessarily need to be spraying at the same time. So I messed with turning this on. By default it uses the primaries up to 97% capacity and will only turn on the secondaries when it needs more than that. Made no immediate change but then I closed the idle screws on the secondaries completely and adjusted the primaries more. Same idle speed but much smoother. I could get it to idle lower but flipping the lights or e-fan would cause it to stall but more on that later.

The other issue I had was timing. I thought I transferred over the timing marks from the original pulley correctly but i never checked to see whether they were correct before pulling apart the motor. The REPU guys online said the pulley resembled a 1st gen RX7 and not any of the REPU pulleys. I honestly have no idea what those marks represent but it’s clear they are not where the truck likes to fire. I had the timing so messed up the trailing was firing before the leading. Long story short, i fixed the spark split, verified it with a timing light, set the timing table to mimic the factory distributor, 10 degrees at idle, and bump to 20 degrees by 1750 RPM, then I adjusted the crank angle sensor until it ran smooth. I may be off by a few degrees but it’s good enough for now. The timing light shows my marks a good 20 degrees off from the pulley marks. The only explanation i have is that the CAS is off a bit or

Adaptive tuning, aka self learning. The ECU for the longest time would not update the fuel map cells automatically. It set a few at high vacuum signal but that’s it. Turns out I discovered a bug in the software. If you check off “arbitrary spacing” for the map, for example changing the RPM gap between cells, adaptive learning does not work and it just updates random cells. I didn’t need arbitrary spacing. It was something i experimented with when i was messing with a TPS tune way back. It took me a long time to discover this...but now that it’s resolved I can successfully use rapid tuning to quickly set a base map and slow converge to have it learn as i drive. The values aren’t perfect, but they help identify high and low points that I can then use to smooth out the curve.

Idle Control. Now that i got the truck driving and once rolling pretty much logging 13.7-14.7 AFRs I decided I would install a IACV. Jenvey supplies the following diagram, but as I learned previously, my particular throttle bodies will not work in this configuration.



The throttle body pictured has a set screw to close off the idle adjustment above the butterfly while mine does not. That means if you replace the lower idle adjustment needle with a vacuum port it will allow too much air to pass through leaving me with an idle above 1,500 RPM. So instead, I’m working on this...

First, I moved the brake booster line back to the rear rotor secondary port. This is where it was from the factory and i saw no improvement in braking with it connected to my new vacuum ports via then vacuum manifold.



Next I am going to put the IACV below the air filter so i don’t need a separate filter for it. I have some additional fittings coming in this week to complete the project. The hardest part will be running the extra wires through the bulkhead Connector, not looking forward to that again.



As for the fuel in the vacuum lines, typically it can be attributed to a vacuum leak. Although I’m wondering if in my case it is related to the pulsing of the the rotor intake cycles, the combined vacuum signal from the manifold, and the fact my injectors are ahead of the ports where I sense vacuum. In other words, gassy air can pass between the ports and linger there...and what I see in the line going to the MAP sensor is just condensed fuel. I was running rich for a while. Now that i have it running better and I’m about to install a IACV that will allow clean air to be pulled through the vacuum manifold, I’m hoping the situation will just resolve itself. Also, I think I’ll start revving the motor when I shut it off to clear out the intake of any gas.
 

bazooka joe

No words
Moderator
Contributor
V
2016 Miata Club, 2014 CX5 Touring, Honda 1976 XL 175
WOW! What a project! Ya that revving of the engine, before you shut her seems to make sense
 
V
'86 RX-7 Base, '79 RX-7 GS
Hmm, not sure how a vacuum leak could cause what I'm getting but I'll take a look. I wonder if you're onto something about the combined ports thing. I have a very similar setup and for some reason my rear rotor seems to be more susceptible to having fuel in its vacuum lines. The line going to my FPR is also significantly more contaminated than the lines going to the brake booster and MAP sensor.

What are you targeting for idle air-fuel ratios? Mine doesn't like to go any higher than 13.5:1. I also have to keep my idle at 1000-1100 RPM to keep from stalling like you mentioned. I don't have any idle control though. Mine will also easily cruise around stoichiometric ratios but I was told to target a little richer. I try to stay at 14:1 or lower depending on load. How's your transient throttle? That's something I'm still trying to sort out. It's the initial tip-in that's still getting me. No matter how many changes I make I can't seem to get the fuel to respond more quickly.
 

chuyler1

goes to eleven
Moderator
V
2013 CX-9
The places I found fuel were low areas in the line where the evaporated fuel can collect. On the REPU the brake booster line loops up high before dropping down and I wonder if that was a factory solution to the same issue. See if you can shuffle around your likes to ensure any liquid in the lines will roll back down to the intake.

One other note, I didn’t bother running a line to my FPR although I may consider it in the future. I just set it at 45psi and left it. I’m not running a turbo so it’s not like i need to up the pressure when boost kicks in. Although dropping the fuel pressure a bit at idle my help eliminate the fuel pump whine.

I was targeting 13.7 at idle, 14.7 at cruising loads, and 13.7 at wide open throttle.

Transient throttle (Predicted MAP on Adaptronic), aside from that initial blip of the throttle off idle mentioned before, was pretty good. I could probably spend more time with it but it didn’t prevent me from driving the truck. Forget about initial tip in for a moment and get the rest of the table right. Cruise around in 2nd or 3rd and get the RPM low. Then tap the throttle and hold it at each of the % values in the table. If it bogs, increase it, if it starts surging forward more than you’d expect for that amount of change, decrease it. When you start getting close, data log to see how the AFR is affected. If it initially handles the transition fine, but then falls on its face, the other settings can help dial it in. I bumped my transition time from 150 to 300ms to provide more time before it went back to using the MAP reading. There are a bunch of articles on Adaptronic’s site on how to adjust the async settings which i will probably start messing with once I sort the idle.



Speaking of idle, it turns out the GM style 6-pin IACV I have wont work with the 440d ECU unless I get a controller board. The board from Adaptronic is more than the cost of a simple Ford 2-wire IACV so I just ordered one of those from a Lincoln town car. One wire is +12v and the other grounds to the ECU. The GM style needed its own power and ground, and 4 aux outputs on the ECU. Pain in the ass.

 
Top