[Phdv61]

And for those interested, an idea of the "ancestors" of MAGIC BOX from top to bottom

First design without RPM acquisition. I tried to do what the JAYCAR unit used by many of you was doing, but not good enough as not accurate. But VNT control was OK. So I added a small PCB ( 2nd below ), with an analog barometer (at that time, the digital one was giving me headaches), and the necessary stuff to acquire RPMs.

Ready to go for industrial (3rd down). First time right. Digital barometer ok. and Bluetooth communication.
But being not so good at soldering, and thinking of sharing my findings with other Patrol owners, I went for a SMC design. Now running in my car.

In the meantime, I progressed on setting software, replay software, and firmware. Did hundreds of hours of test driving to explore and understand all limits, and explain all what I was discovering step by step in order to provide fixes for all.

In between, I burnt components, my hands, bumped my head against the wall, did not sleep much...
Classic. Nothing to worry. But so much fun !

 

[JakeWild]

I'm following with some excitement, because I've always believed that the VNT turbos we have once we converted to the manual turbo control (dawes or Tilix valve) aren't being used to the full capability the turbos. I like the idea of finely controlling the veins for the driving condition. My newly rebuilt ZD30 DI has the Eclipse turbo, remap and soon to have hi flow intake, front mount intercooler and +50 injectors and I'm thinking this maybe the icing on the cake.

 

[Phdv61]

What are you running with at the moment?

• valves only
• ECU + valves ?

How many ?

I believe that @Bidja is the one who has gained the widest experience with -ECU-controlled-actuator-compensated-with-valves, but you are right to say that it is not yet the optimum setting, as you do not control ACCURATELY the level of vanes opening and closing at all times. In addition, the "best" opening of the vanes is not linear with boost. So no pneumatic arrangement whatsoever can precisely deliver that.

I have been arguing for a long time with some of you about keeping the vanes closed for too long with their dawes acting as an ON-OFF button, building boost and back-pressure, limiting on top the amount of hot (compressed) air swallowed by the cylinders. And even if the Tillix brings an apparent improveme,t.

Since, by now, most of you also know the effect of 0.25mm on the actuator rod length, you will easily understand that only electronics and software ( like what an ECM would do...) can provide the required rod positioning accuracy for a given Patrol configuration.

Our ECMs were optimized at the time of their design for their Stock configuration, power/ torque. Any mod, and they get lost. And you are left with work-arounds.

 

[JakeWild]

What are you running with at the moment?

• valves only
• ECU + valves ?

How many ?

I believe that @Bidja is the one who has gained the widest experience with -ECU-controlled-actuator-compensated-with-valves, but you are right to say that it is not yes the optimum setting, as you do not control ACCURATELY the level of vanes opening and closing at all times. In addition, the "best" opening of the vanes is not linear with boost. So no pneumatic arrangement whatsoever can precisely deliver that.

I have been arguing for a long time with some of you about keeping the vanes closed for too long with their dawes acting as an ON-OFF button, building boost and back-pressure, limiting on top the amount of hot (compressed) air swallowed by the cylinders. And even if the Tillix brings an apparent improveme,t.

Since, by now, most of you also know the effect of 0.25mm on the actuator rod length, you will easily understand that only electronics and software ( like what an ECM would do...) can provide the required rod positioning accuracy for a given Patrol configuration.

Our ECMs were optimized at the time of their design for their Stock configuration, power/ torque. Any mod, and they get lost. And you are left with work-arounds.

I'm just running a single Dawes and the remap.

 

[Phdv61]

What a tremendous error !
Go to the relevant threads and talk to @Bidja.

 

[trev zd30]

What a tremendous error !
Go to the relevant threads and talk to @Bidja.

Hey Jake,

I'm running the 4 valve arrangement that Phil suggested. Come and take it for a drive one day if you want to compare to single dawes. Much smoother boost control.
As you prob saw I'm one of the lucky ones getting the majic box to test. Happy for you to check it out when I get it in👍

 

[Phdv61]

As explained in a PM to Trev, and in the documentation provided,
Magic box can adapt and make use of whatever 0-5V MAP sensor you may wish to use, to overcome the stock MAP sensor limitation to 22+psi (1.5 bar) relative. Make sure you select an ABSOLUTE SENSOR ( measuring negative boost RELATIVE ) and NOT a cheap chinese 0-30 PSI RELATIVE, or you would loose the ability to adapt to all atmospheric conditions.

At sea level, atmospheric pressure varies between 95kPa and 105kPa depending upon the wheather conditions.
In altitude, it can go down to 70kPa at 3000m. ( Keep in mind the 0.25mm actuator "rod effect" ).

The stock MAP sensor goes from 133 mbar absolute (0.5V - 0.48 psi ) to 2533 mbar absolute (4.5V - 36.74psi). MAGIC BOX comes with the right setting for this sensor, but this can be easily modified if you fit, or want to use, another sensor.

 

[JakeWild]

Hey Jake,

I'm running the 4 valve arrangement that Phil suggested. Come and take it for a drive one day if you want to compare to single dawes. Much smoother boost control.
As you prob saw I'm one of the lucky ones getting the majic box to test. Happy for you to check it out when I get it in👍

PM sent.

 

[Phdv61]

Sorry Guys,

The first three boxes for Di, and the same for CRDs already found their future owner.

I will build the Di boxes next week, and wait to assemble the CRD 's until I know they too acquire the RPM signal properly.

I will now order more components to be able to build a few more boxes without a two weeks delay.

Enjoy your summer week end.
Here, we never had so much snow, but all ski resorts are still closed due to the pandemy.

 

[Nikolay]

how much is?

 

[Phdv61]

Hi Nikolai,
have a look back a few posts.

 

[Phdv61]

The two Magic Boxes eventually arrived yesterday in-country.
We'll see next how they install and behave on CRDs.

In the mean time, I cleaned-up the install and operation document with the help of @Bidja and @trev zd30 so that it can be used for both 3.0Di and 3.0CRD.

I also improved the "cosmetics" of the setting and replay software, and created a detailed dated log, to keep trace of all setting actions performed, and be able to provide further "distant analysis and support" if need be.

 

[Bidja]

Bidja Update.
@trev zd30 and myself received our Magic boxes from @Phdv61 and both have successfully paired the the device via BT to our 64Bit computers.

Need was to pwr up the box on the bench (no sensors connected), I used a 240vAC - 12v DC 3A transformer and connected 12v +ve via 2A fused patch wire (supplied) + earth connection and used a WAGO twin connector for hook up. All powered up fine and Paired.

 

[Phdv61]

John and Trev are progressing their install on their respective CRDs. I just set them the MAGIC BOX Android App for their smartphone. The App will not allow to set parameters when driving ( that would be far too dangerous !) but will allow monitoring of / Boost / MAF and TPS voltage / VNT control / on your dash.

Same color code as the PC software. Boost in Bar here.
Green : Boost.
Blue : MAF voltage
Grey : TPS voltage
Yellow : VNT control ( 20% to 60%) vanes fully open to vanes fully closed

 

[Phdv61]

Running...
4th gear. 63 km/h.
I just forgot to shut the back door properly...
I took the photo with my iPhone. Not so easy. See my finger reflected on the Samsung screen...

That old phone (10+ years old) is only capable of refreshing its display 4 times per second. Just enough to get a "real-time" feeling.

 

[Bidja]

Phil, received your email of the BT paring app (I guess) for the android device display as above:

is there any special instruction for uploading app to device.
Will do bench hook up of Magic Box tomorrow and give it a run_is this worthwhile?

 

[Phdv61]

Hi John,
I sent you an email with the app in psi as an attachment (.apk).
Trev apparently had difficulty to download. Will sort that out.

 

[Phdv61]

Hi Guys, it's been a long long time without news, but John, Trev, and myself got several personnal problems to sort out which delayed the set to work and testing of John and Trevor's respective magic boxes.

Hopefully, we'll be soon in a position to provide you with some feedback.

I can confirm dawes + needle is definitely not the arrangement you want to keep. Most of us knew that already.

The thing is that, no matter how many valves you add, even if you can improve your vanes/air management, you will never get the degrees of freedom required to do things right, as the optimum rate of opening at all times depends upon the boost already built, and the load requested. This is something you can't set properly with valves.

Getting the most out of any VNT turbo is therefore not as simple as fitting two valves, three, or even four. The way your turbo vanes have to be managed is linked to your Patrol configuration, turbo brand and model, exhaust diameter, inlet, rod length .... etc, and there is no reason for the rate of optimum vanes opening be linear with boost. In fact, it is not. At all. If you want to really maximize airflow, improve your patrol driveability, and get rid of peaks and drops of boost once for all, stay tuned. You are in the right place.

More detailed information to come.

 

[J0SHMAN]

Thanks for the update, and good luck. I'm watching this with a keen interest.

 

[Phdv61]

So am I !

The good news for me is that only two sensors are different for the 3.0 CRD ( RPM and TPS ), compared to the 3.0di and their proper acquisition is now validated. It was my only real concern as I did my adaptations mods "blindly".
MAF and MAP are identical to my 3.0Di. And the EV control is exactly the same.

The physics/mechanics/electronics being what they are, I am now fully confident that the end result on the CRD will be as good as what I get on my 3.0Di.

John has a different turbo than Trev, so their settings will be different, and different from mine obviously. This is the beauty of software and "maps". You can get what you set them for.

The real "difficulty" is to forget a bit about boost as such. Boost is not all. Boost must be kept "under control" at all times. Only by adapting the "turbo map" will you maximize airflow at all times ( and therefore torque/power ) whilst keeping max boost under control and below the maximum you have decided to "authorize".

Trev has demonstrated also that reducing the MAF voltage when idle can improve fuel eco when idle by 40% ( the CRD has an OBD2 port, unlike my 3.0Di).

In my case, with a 3" exhaust and an EGR blocked, my MAF voltage when idle was 2.15V. I have reduced this down to 1.7V ( which is - 0.45V !), ONLY when idling, and my Patrol loves it.