yavor

Hi Everybody,

I'm new to the forum and I'm also new to the Miata world.

A few months back i decided to buy a 1998 NB 1.6 110bhp with relatively low milage and no rust. I bought it with boost in mind.

My plan is to carry over my current DIY turbo project (Opel Vectra 2.0l 8v making 180bhp and 290Nm @ 0.65bar on LPG) into the NB. The power goal is ~150 crank hp and ~200nm crank torque. The budget traget is minimal

The engine
I plan to use the stock 1.6 in almost stock form. I'm going to change to a 323 GTX head gasket to be able to handle the extra cylinder pressure safely. Mine is leaking in the corners anyway. Bottom end would stay stock.

The Turbo
The plan is to run my TD04HL-13G (coming form a T5 Volvo) which should be able to achieve my goals at about 6~7psi (0.4~0.5 bar)

Fueling
The plan for fueling is somewhat alternative. I'm going to use the stock ECU with LPG injection as piggiback. Petrol would be used for startup, warmup, limp to the petrol station. LPG as the primary fuel. The LPG computer has fuel tables capable of scaling fuel to meet AFRs under boost. This has worked well in the Vectra for over 10k km.

Piping and ducting
Exhaust manifold, intercooler piping and exhaust are going to be hard core DIY. This has worked in the Vectra
The intercooler would be a chinese universal unit with maybe 6-7l of volume. It should be big enough for this application. In the Vectra i used small a Pajero 3l unit which was a little small in hot days.

Now i have a few questions that are rather specific and would require your help.

Ignition
at 7psi with a lager intercooler and LPG (108 octane) would i be safe to not touch timing or if i need to, I probably not need more than 2-3 degrees of retard. If i need to retard a bit are there any DIY low cost options.
In the Vectra i had a jumper on the ECU capable of changing between 95 and 91 octane (EU spec). It also had a resistor mod which allowed extra 5 degrees retard. It was quite happy running the 91 mod and it didn't have any knock until like 5500 RPM (after that i was going out of the compressor map and probably producing large amounts of heat)
I've read that in NA miatas you could rotate the CAS and allow for 2-3 degrees retard but in the NB there are Crank and CAM sensors which do not allow for any movement or adjustment.
I read that some people use intake air temperature or coolant temperature modifiers to achieve a few degrees retard.
The question here is whether the stock NB 1.6 ecu has a temperature retard feature for either IAT or ECT.

Turbo Oil feed
as far as i know the TD04 requires a lower oil pressure to run normally. I know that i could use an oil cap in the block right next to the oil sender and a restrictor on the turbo but such restrictors are not readily available in my country. I've seen there is an oil cap at the back of the hear and that there is a restrictor built into the block to head oil gallery. So in other words the oil pressure in the head/the back of the head should be lower that the oil pressure in the block/next to the oil sender.
The question here is has anyone measured oil pressure at the back of the head. Would it be safe to run the turbo feed off that heat port.

Clutch
I'm planning the car for street use and i wanted to know if the stock 1.6 200mm clutch would be able to hold the ~200Nm torque. If not what would be a relatively cost effective clutch upgrade. I've seen some no name brands like F1 Stage 1 or 2 selling for as little as 120~150$ Does anyone have feedback on those.

Cooling System / Coolant reroute
I've read about the factory coolant flow issue of miatas and the need of coolant reroute. I've also ready that Mazda supposedly fixed it in 2001 in the VVT 1.8 by changing the head gasket coolant hole pattern. I already checked that my current and the 323 GTX head gaskets have the same pattern as the fixed 1.8.
The question is have to looked into this aspect and do we really need the coolant reroute for as little power as 150bhp.



Wow the thread turned out a little long but the Miata is still quite unknown to me so i have a lot of questions.

Any help would be appreciated.

Thank you.

Lexzar

iTrader: (2)

I know you might be against it, but even for low boost stuff, I'd still want a standalone ecu like a used MS2pnp or similar to control fuel and timing.

I know the clutch will probably die. And the coolant reroute won't be needed.

shuiend

iTrader: (24)

I understand that you want to do this on the cheap. I would skip the no-name brand clutches. I have seen to many fail, way to quickly, to be worth it. Yep you might save $200 up front, but after having to swap the clutch multiple times and downtime while waiting, along with shipping costs to Bulgaria I would shell out for a FM1 clutch.

Also you have pretty good english for a Bulgarian. You also made a decent first post asking questions. LPG is not typically used in the US for cars, so you will be mostly on your own for using it as a fuel source. But it sounds like you have done this before and have some experience.

adryargument

iTrader: (1)

Try the oil off the back of the head, too much oil will simply give you blue smoke on acceleration. Then build your own restrictor if required - considering your DIY piping.
i.e. fill a bit of steel pipe with weld and drill a hole / shove a 0.9mm welding tip into the tube / drill an end cap or blockoff pipe... I'm sure you can use your imagination.

The 1.8 clutch slips at 7psi on the same turbo. I'm sure the 1.6 is rated for less.

Surely the 01/05 head gasket is useable? Or is the increased pressure your talking about to do with LPG?

yavor

Thank you for the inputs on the coolant reroute and clutch.
As far as the clutch goes I'll probably go with the Excedy Stage 1 1.6 as it is easier to find in Europe and also a direct fit. For the FM clutch i would need to find a 1.8 flywheel.

These are mods I'll probably do later on as the project goes.

Now the only thing left to find out is Ignition retard and oil pressure.

For NA I've ready that it does not change ignition based on IAT but cannot seem to find any infos for NB on either IAT or ECT.

yavor

Thanks for the inputs. I thought about welding the oil feed fitting on the turbo side and drilling it. I've done something similar with the restrictor pill on the vectra boost controller. If i can't figure out the oil pressure, i'll make the restrictor.

BTW i have a fuel pressure gauge from an AFPR could i use it to measure the pressure off the top and bottom of the engine. I mean can i use the fuel pressure gauge to measure oil pressure.

For the head gasket it's not really and issue with LPG, i had some bad experience with the vectra where i blew the na gasket at around 0.4bar. I then swapped to a c20let (factory turbo from same period) HG. Every since it has see quite a lot of abuse and boost creeps up to 0.8 and still holding strong.
This is why i prefer a factory grade turbo gasket. Something in the materials is different and they are stronger. This gives me a piece of mind.

shuiend

iTrader: (24)

Get some various sized pop-rivets and use those as the restrictor.

I am not sure if the 1.6 in the NB uses the organic HG or the MLS HG. I would swap to a MLS HG if you can.

As for the timing, you could possibly get a 5xRacing timing wheel and it might work. I have never worked on a 1.6NB before so not sure what exactly is different then what we have in the US.

yavor

The HGs on the NB and 323 GTX are both Organic. MLS would be hard to find for 1.6 and i believe it would be an overkill for my power goals.

5xRacing trigger wheel might work. My only concern is that it only changes crank angel and not cam angle as well. This could create some issues for the ECU. Changing the base timing would also mean i lose off boost performance.
If changing the trigger wheel angle is enough, I've read i could drill out the Crank sensor holder and move the sensor itself instead of the trigger wheel.
If I'm able to change timing with resistors (electronics) i could use a pressure switch and make it boost dependent just like Greg from The Car passion channel did with Broke and Boosted O2 override.

ryansmoneypit

iTrader: (6)

I agree on the welding tips for oil control. they are pretty accurate and easy to source in different sizes.

DNMakinson

iTrader: (3)

See https://www.miataturbo.net/diy-turbo...-supply-75631/ for some oil pressure readings. There are questions about if I ran the test correctly, but I give the method and the results, so you can decide if my data meets your needs, or not.

I use the back of head port, a TD04, and a 1mm restrictor that is built into the Turbo oil input adaptor. About 10k miles boosted with no discernible issues, but full disclosure, I have not had the valve cover off.

Joe Perez

iTrader: (8)

The TD04H-15G was a very popular turbocharger for 1.6L Miatas in the US for many years, as it was supplied in kit form by Greddy. I had one myself, and fed it oil from the port on the side of the block with no restrictor.

Recommendations about needing to restrict oil supply generally apply only to ball-bearing turbochargers. Journal-bearing units require higher pressure, and are generally sufficiently restrictive on their own not to require an external restriction.

yavor

Thanks for the infos on the oil feed. I was looking exactly for a test like DNMakinson did. In the head the pressure appears to be a little lower than what i expected. I'll use the port on the side of the block. I'll try restrictorless. If I have exhaust smoke I'll weld and drill the turbo in fitting to a 1mm restrictor.

Meanwhile Saturday i took the Vectra apart. Here are some pics of the various pieces and the turbo. The old exhaust is 60.2mm stainless, I'll reuse it as charge pipes and downpipe as it has some usable bends and elbows.

It turns out i could reuse the exhaust manifold but it would be a little restrictive. Between ports 1 and 4 the distance is 308mm which is exactly the same in the B6. The issue is that Vectra has round ports 32mm each, where the B6 has oval ports 31.1x49.9mm.

The new one would probably be the same design but 48.3mm the whole way around.

Here is how 60bhp look like
Turbo, old DYI exhaust manifold, boost controller, and some charge pipes
Had a small leak from the oil feed.
Turbine side looks clean after 10tkm
Compressor looks pretty clean.
One of the following days I'll take the Miata Exhaust manifold off and do a quick mock up to see which goes where.

I hoped to be able to mount the turbo below the manifold but space could get tight with all Hydraulic pump, water pipe and steering column all occupying the same space. Side or top mount would probably work without issues.


By the way, does anybody know what CC a stock NB 1.6 injectors. They are slim design with blue body. I cannot seem to find info anywhere. Now some say they are 200cc others 235cc.
I've done a air/fuel mass calculation. It appears for 110bhp 200CC are sufficient at about 75~80% duty cycle.
could it be that 110bhp 1.6 has 200cc and 125bhp 1.6 has 235cc injectors.
The reason i ask is that I found a dirt cheap 3:1 FMU.
In theory I could be able to get her going with 235cc injectors a 3:1 FMU at 0.5bar of boost.
This would mean a fuel pressure of 4.5bar and new injector size of 281cc.
This would result in 156bhp @ 80% duty cycle and 0.55 BSFC.
Has anyone run a similar setup or can anyone prove me wrong or right.

Thanks

yavor

Hi Everybody,

it's been a while but the project is still going
I've had my up's and down's and almost put the car for sale a few times... but am still going
I managed to fix most of the gremlins that haunt a 19 year old Mazda and have fresh ideas of going turbo.

I decided to scrap the LPG thing for now, not that it wouldn't work, but because the LPG tank has to be 20liters to not take up the whole trunk. This would have given me a range of about 100km of spirited driving.
I though about megasquirt, FMU, larege injectors with MAF modifiers and what not, but all of those options are either too expensive or too unstable.
So a few days back i had a brain fart of trying a 5th injector setup. I've given it some though and figured i could use an Arduino to control it.
The idea sounded so crazy that it gave me inspiration to move forward with the project. This setup could be done for less than dirt cheap

I managed to write and test the Arduino code on a virtual machine and will try it out on a real board the following few days.


Today i took the car apart and did some testing. It turns out the Vectra manifold mounted upside down fits the Miata perfectly. It's only about 1cm longer than needed so i guess i'll cut the runner on cylinder 4 and weld it back shorter.

Here are some pic of the current mock up:




I'll post some update once i get the arduino stuff tested

yavor

Hi Everybody,

today i bought the arduino and started testing / doing a proof of concept. I had a little issue reading the RPM input which i didn't in the virtual machine but now it's fine. Here is the first shot of the brettboard:


The button represents the RPM pulses. The Pot represents the MAP sensor and the voltage it provides in different load conditions. The LED on the right represents the Injector and show injection pulses and pulse duration. One of the three LED's on top will eventually become an O2 clamp if i see the need for it. For now I've put them in for fun. They light up consecutively as boost rises 0.1, 0.2, 0.3 bar.
On the bench all works fine. Tomorrow i will test the setup in the car and see if I'll be able to read RPM normally. I'm still wondering whether to get RPM signal from an Ignition coil or from an Injector. I guess the coil would show 6000 impulses at 6000 rpm(wasted spark) where the injector would show 3000 impulses (if running sequential). I figure both would work, it's just that one would have a 10ms injection duration limit and the other 20ms which is easily adaptable in the code.

yavor

Today i tested the setup in the car. I was able to read RPM successfully. In the Diagnostic socket there is an IG- pin which gives direct tacho input so 2xRPM. I ended up using the Ignition coil input. Here a short video of the test:
Winding the POT changes the boost readout (the 3 LEDs light up) and the injection duration. Off boost produces no injection pulse. On boost produces injection pulses. The higher the boost, the longer the injection pulses.

In the afternoon i realized i have a MAP sensor from the LPG system off the Vectra. It's a 7bar map but still it's able to tell the difference of me blowing lighter or harder in the boost line and most importantly it's free

Here is a short video of the MAP test:
the 3 LEDs light up according to boost.

Next step would be to transfer the brettboard to a more permanent circuit board and make the injector driver (mosfet transistor circuit). I found some old LPG injectors which i would use for dry testing.

Pepovr

Hello, kolega. What city are you from?

yavor

Hi I'm from Sofia

yavor

Hi Everybody,

it's been a while and a lot of things have changed. The last update to the project is that i sold the Miata, the Vectra and everything turbo and performance related. I started spending too much time in the garage and got on my wives nerves, so I had to do what had to be done to restore the piece :(

So I wish you all the best and hope to hear from you if i ever have the chance to buy another Miata.

Cheers

g04lucas

so question about the clutch remark, when you say fail way too quickly...like around how many miles?