LPG beim Thesis

[donthema]

Gents, I think the Thesis module adaptively learns and adjusts the mixture to get the correct air / fuel ratio. So if you set the LPG unit to lean, the Thesis module will simply set longer injection time to compensate and get the correct AFR again. Unfortunately it will use these values also if you cold start with gasoline. Due to the compensation, the injection is longer and with gasoline it's now rich. So you can get adverse effects.

[vlad47]

LPG works a bit differently. The wires from thesis control unit to petrol injectors are disconnected and lead to LPG control unit. LPG control unit is injecting the gas according to the values from petrol but they are not the same. LPG control unit corrects them to LPG. Reasons are more, different gas pressure, different gas injectors size etc. LPG control unit also simulates for the thesis control unit so it believes everything is OK and petrol is used. Some LPG control units uses also information from lambda sensors and my new one will have also access to all data through OBD connector. All datas are used for the best settings like pressure, injection times etc. You are right that thesis tries to keep the best mixture but at first place he can never do that because he thinks you are injecting petrol (petrol has different energy per unit) and the second is that these injection times can be corrected but LPG control unit doesn't set them exactly, it only considers them. Another fact is that LPG control unit is trying to get the same injection times as driving on petrol.
All these facts are for LPG control units with learning function.
Starting the engine has nothing to do with previous datas and settings. Values are set from factory for all conditions. Older settings can't be used because the engine was hot before and now the engine has no information from lambda sensors because they are cold even if they are heated, but they start working only after a few seconds.

[vlad47]

Well, I've finally finished first part of the LPG project, the injection. As you can see on the pictures the gas is injected through original petrol injection space by special inlet plugs, which forced me to have petrol injection rail higher, what pressed me to have the air inlet pipes higher what get me to put also the air head higher. Special insertions were used for all. Gas injectors are as close as possible to the valves. This method was used for the best possible efficiency gained by the most precise injection. For this was also chosen the injectors type. Pipes are orange because they are not normal gas car pipes but gas pipes for welders, they are ten times more stronger, but more flexible. As short as possible of course. Each second end goes to the filter with integrated distributor and temperature sensor.
All took 48 hours clear time.
changes:
- consumption eventually less then petrol
- more power and torque than on petrol, even more than I predict
- engine running more smooth than on petrol
- lambda sensor characteristic seems nicer than on petrol
My project is still not finished, today I've ordered two new reducers (one is able to power 380HP), the point is to achieve stable pressure, what is possible only in ideal conditions, not in real, that's the reason for such overequipped decision.
the MAP sensor will be changed too.
I will follow informing, I believe all would be done at the end of next week.

www.lanciathesis.eu/pictures/lpg/index.php

[starbiker]

Very interesting report, Vladimir.

You said "consumption eventually less then petrol", will be interesting to see how you get more kilometers from LPG which has less energy content than petrol....

I have recorded my LPG-consumption for the last 4 years and compared it to the average consumption which is shown in the Connect's Trip-computer: roughly 25% higher.

Cheers.....Ralf

[vlad47]

I'm measuring the difference between trip and reality and it's 0%. It depends on the lpg control unit, but if I were you I would exactly look at it, my consumption used to be 4-6liters on one full tank more than the computer showed, but the problem was lpg leaking. After many tests I found out (tested on two different cars) that you are not able to fasten it 100%. Also if you do, after some weeks it starts leaking again. So, disassemble the pipes use special engine sealant for high temperatures (red one). After work done your all connexions will be 100% fastened. I guarantee it, because after some weeks when I tried to disassemble one pipe again, I saw how strong does it hold. 25% is extremely high number, there must be something wrong.
Not to forget, I know you are right that one liter LPG has less power than one liter petrol but you must consider it's measured in liquid state. There are too many other things to consider. There are cars (with direct petrol injection) where LPG is injected directly in liquid state. If done so, the engine has more power and less consumption. Why? It's too easy, LPG has -400 degrees in liquid state, the colder the mixture is, the more oxygen is there and the better is the burning efficiency. Using LPG injection in gaseous state has some more things you have to consider, for example the reducer efficiency depends on the water temperature, which is used for changing states from liquid to gaseous, higher the temperature, the more gas it can produce. We don't have to forget also the precision, for example usually gas injectors are very far from the engine and there's gas staying in the pipes, mixing with not sucked air. You know these all things are possible, if not our cars would still consume 25liters petrol like 50 years ago. It's all about efficiency and problem is that LPG is too young and has many many things to improve.