BennehBoy's D2 adventure

Subbed, great work mate.

Eh up.

Not been doing much with the Disco for a bit other than a very low amount of miles (less than 3k a year for last couple years). I'm doing a lot more now and the last fill up was almost 150 quid so....

I've dusted off my gauge system again. I want to read the fuel injection quantity from the ECU so I can display a fuel economy figure calculated from that and RPM's.

Basically if I drive this thing like I would normally I get 320 miles to the tank !!! 320 reported miles, the tyres are 5% bigger circumference than stock so the mileage is under reported.

If I drive like miss daisy, I can get 470 to the tank - I think that having a big flashing MPG or litres per 100mk figure on the dash will help remind me to keep the foot steady with 2022's ridiculous fuel costs.

Anyhow, I've managed to get my electronics talking to the ECU, I just need to do some code tidy up now - and probably finish off adding the other physical instruments that I've been putting off for a while : R380 cooler, oil temp & pressure senders, LT230 Oil temp, Engine Oil temp & Pressure.

I also need to finally fit those blooming Sumo bars

Here's a photo of the bench test system:


I had been gifted an ECU previously that turned out to be toast, I snagged this one in the pic for 60 quid from a lad locally and it works a charm - I've got a logic analyser hooked up so I can simultaneously sniff the K-line & the Serial activity driving it from the microcontroller - which basically means I can figure out wtf is going on and that I'm getting stuff out of the ECU.

At the moment I'm wrapping this all up with the code that reads all the extra analogue sensors, but I might just release this as a stand alone fuel consumption system - be a doddle for anyone with a TD5 to use it then, and probably a lot cheaper than the competition. I think there's about 20 quid in parts for the ECU side of things - gets a bit more expensive when the analogue sensors and cabling/connectors are factored in.

Bennehboy said:

Eh up.

Not been doing much with the Disco for a bit other than a very low amount of miles (less than 3k a year for last couple years). I'm doing a lot more now and the last fill up was almost 150 quid so....

I've dusted off my gauge system again. I want to read the fuel injection quantity from the ECU so I can display a fuel economy figure calculated from that and RPM's.

Basically if I drive this thing like I would normally I get 320 miles to the tank !!! 320 reported miles, the tyres are 5% bigger circumference than stock so the mileage is under reported.

If I drive like miss daisy, I can get 470 to the tank - I think that having a big flashing MPG or litres per 100mk figure on the dash will help remind me to keep the foot steady with 2022's ridiculous fuel costs.

Anyhow, I've managed to get my electronics talking to the ECU, I just need to do some code tidy up now - and probably finish off adding the other physical instruments that I've been putting off for a while : R380 cooler, oil temp & pressure senders, LT230 Oil temp, Engine Oil temp & Pressure.

I also need to finally fit those blooming Sumo bars

Here's a photo of the bench test system:

I had been gifted an ECU previously that turned out to be toast, I snagged this one in the pic for 60 quid from a lad locally and it works a charm - I've got a logic analyser hooked up so I can simultaneously sniff the K-line & the Serial activity driving it from the microcontroller - which basically means I can figure out wtf is going on and that I'm getting stuff out of the ECU.

At the moment I'm wrapping this all up with the code that reads all the extra analogue sensors, but I might just release this as a stand alone fuel consumption system - be a doddle for anyone with a TD5 to use it then, and probably a lot cheaper than the competition. I think there's about 20 quid in parts for the ECU side of things - gets a bit more expensive when the analogue sensors and cabling/connectors are factored in.

Click to expand...

Feck I am impressed!
I don't do electrickery and your pic looks like the sort of mess my other half would despair of and tell me to "Get tidied up!"
I just try and drive at about 2000 revs to stay the MPG side of sanity. But it is slow.
Last September I had to drive 550 miles at 50 odd miles an hour, boy did i save some fuel!!!

Got the prototype working in the car.

RPM & Injection quantity being read, but I now need to update the code to display a calculated consumption figure.

Happy that the electronics is all sound though.

Can someone with half a brain help me work through my maths

I want to be able to calculate a litres per 100 km figure (probably also MPG once I have it straight in my head).

In order to do this, I need to know the injected fuel per stroke, the vehicle RPM, and the vehicle speed.

There are probably a lot of ways to go about this and no doubt I've probably picked the hardest way or got it completely wrong

First I need to get to kilograms consumed per hour...

= injection quantity in mg * rpm * (number of cylinders / number firing per revolution) * minutes in an hour / milligrams per kilo
= injection quantity * rpm * (5 / 2) * 60 / 1000000

At idle my car is injecting 22mg of fuel at 750 rpm, plugging that through the equation comes to 2.475 kilograms per hour - assuming I'm reading the correct value from the ECU!

Next we need to workout what that is in litres. So average density of Diesel is ~ 0.85, obviously this varies with temperature and I could calculate it from reading the fuel temp, but it's only meant to be a guide not exact.

Therefore litres per hour = kg/h * 0.85

So at idle for me that's 2.475 * 0.85 = roughly 2.1 l/h

To see what that equates to per 100 km, and I know this is contrived because the vehicle is stationary, but I'm just using the above figured to sanity check my workings:

100/vss gives us the number of hours it takes to travel 100km

So if we multiply our Litres per Hour by this figure we get the number of Litres per 100km.

Assuming the vehicle was in gear and moving at idle and the injection quantity didn't go up due to load, and assuming I got 5kph for that rpm (utter bollocks figure), then:

100/5 = 20
20 * 2.1 = 42l per 100km
Probably not far off the truth at that speed

Do my working outs look about correct?

Bennehboy said:

Can someone with half a brain help me work through my maths

I want to be able to calculate a litres per 100 km figure (probably also MPG once I have it straight in my head).

In order to do this, I need to know the injected fuel per stroke, the vehicle RPM, and the vehicle speed.

There are probably a lot of ways to go about this and no doubt I've probably picked the hardest way or got it completely wrong

First I need to get to kilograms consumed per hour...

= injection quantity in mg * rpm * (number of cylinders / number firing per revolution) * minutes in an hour / milligrams per kilo
= injection quantity * rpm * (5 / 2) * 60 / 1000000

At idle my car is injecting 22mg of fuel at 750 rpm, plugging that through the equation comes to 2.475 kilograms per hour - assuming I'm reading the correct value from the ECU!

Next we need to workout what that is in litres. So average density of Diesel is ~ 0.85, obviously this varies with temperature and I could calculate it from reading the fuel temp, but it's only meant to be a guide not exact.

Therefore litres per hour = kg/h * 0.85

So at idle for me that's 2.475 * 0.85 = roughly 2.1 l/h

To see what that equates to per 100 km, and I know this is contrived because the vehicle is stationary, but I'm just using the above figured to sanity check my workings:

100/vss gives us the number of hours it takes to travel 100km

So if we multiply our Litres per Hour by this figure we get the number of Litres per 100km.

Assuming the vehicle was in gear and moving at idle and the injection quantity didn't go up due to load, and assuming I got 5kph for that rpm (utter bollocks figure), then:

100/5 = 20
20 * 2.1 = 42l per 100km
Probably not far off the truth at that speed

Do my working outs look about correct?

Click to expand...

MAD!
Engines consume feck all at idle, unless working.
When working they consume far more so yes, the idle figure will not help you with a figure for when it is pulling load.
Why not just fill it to the brim, absolutely to the brim?
Start it, run it for half an hour at idle. Then measure, very accurately, with a syringe or summat, how much it takes to top it up?
It'll take more starting from cold, so you could warm it up first, but they never get very hot only idling, nothing like proper driving.
Please tell us why you are doing this?

OK I think maybe I've not communicated the purpose clearly

This is so that the computer device that I'm using in the car already, can calculate a live consumption figure.

I don't care about the idle figures, I just used them to have some data to pump through the equations to see if I'm on the right track.

I just want to validate the maths.

Bennehboy said:

42l per 100km

Click to expand...

Something is wrong there, 42 liters. 4.2 i would think. I can get into the high 8/100 in a 4.4

Problem is as load request increase does not the injection quantity?
Or is it the diesel just runs the same quantity all the time? especially with a turbo.

I just remember when I was playing with a petrol RX that injector duration was a variable as was boost and fuel pressure. but that was petrol.

Will look at you math but not tonight.

But I do like the fact you have taken this on.

J

OK I shouldn't have included the idle injection stuff at all, that was just me making sure the figures weren't wildly wrong. More interested in whether the conversions I'm doing make sense - I _think_ they're correct, but not many people out there are doing live calculations so most stuff on the internet is about working out fuel consumption after the fact based on miles travelled and how much the tank needs topping back up.

Bennehboy said:

OK I shouldn't have included the idle injection stuff at all, that was just me making sure the figures weren't wildly wrong. More interested in whether the conversions I'm doing make sense - I _think_ they're correct, but not many people out there are doing live calculations so most stuff on the internet is about working out fuel consumption after the fact based on miles travelled and how much the tank needs topping back up.

Click to expand...

duuhhh!
Like me!

Bennehboy said:

OK I shouldn't have included the idle injection stuff at all, that was just me making sure the figures weren't wildly wrong. More interested in whether the conversions I'm doing make sense - I _think_ they're correct, but not many people out there are doing live calculations so most stuff on the internet is about working out fuel consumption after the fact based on miles travelled and how much the tank needs topping back up.

Click to expand...

Can you see your map? and its different sections.

J

marjon said:

Can you see your map? and its different sections.

J

Click to expand...

Not sure I follow, do you mean the fuel map? I'm talking directly with the ECU and reading the actual injected volume of fuel at any given moment, so once this is in the car and not just hooked up to my bench ECU it will read the actual RPM, injection quantity, and Speed to do the calculations against.

Bennehboy said:

Not sure I follow, do you mean the fuel map? I'm talking directly with the ECU and reading the actual injected volume of fuel at any given moment, so once this is in the car and not just hooked up to my bench ECU it will read the actual RPM, injection quantity, and Speed to do the calculations against.

Click to expand...

As I said I don’t know diesels. But know you have done loads to yours.
But does the injection quantity change with boost or load? Or is it constant?

Fuel rail pressure?

J

Ah OK I see, the injection quantity changes based upon driver demand (pedal). So, no it is not constant. I think the map on mine has it inject up to 85mg at full chat, but that is also at like 4500 rpm, so quite a bit of fuel.

Bennehboy said:

Ah OK I see, the injection quantity changes based upon driver demand (pedal). So, no it is not constant. I think the map on mine has it inject up to 85mg at full chat, but that is also at like 4500 rpm, so quite a bit of fuel.

Click to expand...

And if I see it you used the figure 22mg in your calcs. So this can’t be a constant through your time rpm calc?
How does it adjust quantity, duration or fuel pressure? Or both? Can you monitor those and calc quantity?


J

Pressure is constant, the ECU varies duration of the injection firing solenoid, from that it knows the fuel quantity injected. There are a series of maps that combine based upon RPM and driver demand to determine the duration.

There's very good info on here -> https://discotd5.com/ecu-tech/td5-tuning

For my purposes I don't need to know that stuff, I just need to grab the value the ECU thinks that it injected.

Ok.
Yes I would say you have enough. Give it a go.

Will read link later cos I do enjoy this stuff, although not dabbled in years.

Does your setup just give instantaneous number or is over a distance (tank fills/miles)?

can you add a correction factor?
So when you check brim to brim numbers you could adjust if way out. I actually read somewhere in the L322 there is a correction factor than can be adjusted, but wasn’t what I was looking for at the time so didn’t really take it in, as you do.

let us know how it goes please.

J

For now it will be instantaneous, but I'm thinking about how to at least make it trip specific. Going beyond trip ie. retaining the values across ignition on/off is tricky with the device I'm using because it has no internal eeprom (unlike a much slower Arduino device) so I would have to add some more hardware for that.

I forgot to say, yes I'll be baking some adjustment factors in - to be honest these will probably be 'set at programming time', because again, storing any data needs more hardware, and being able to make adjustments needs a menu system and some input (a rotary encoder for example) - I only just removed all that stuff from the code because it was getting way too much feature creep.

An adjustment that would be useful - corrected road speed, VSS is based upon stock wheel size, mine are larger diameter.

Off work this week, so thought I'd finally get round to fitting the Sumo Bars that have been sat in the shed for an age.

Given that my factory steering guard has completely rotted out, and that these bars are solid stainless steel and super strong, I decided to get the angle grinder out and chop the old steering guard off. The TF steering damper and HD bars that I put on quite a few years back are still working but they've been eaten to bits by the salt that gets put on the roads here in winter. I've no clue why any of the manufacturers go for passivate coatings in the UK, it's proper garbage stuff for protecting metals.



Here's how bad the steering guard had rotten:


The old drag link came off easy:


Shiny new drag link on, I've completely slathered all the mild steel parts with copper slip in the vain hope of protecting them from rusting:


Old and new track bars, measured up and set new one to same length as the old:


And fitted, same slathering up treatment:


OME steering damper on with a shed load of copper slip:


Job's a good un.