Converter vacuum line

Post Reply
New member
Posts: 40
Joined: Fri Apr 10, 2020 12:07 pm

Converter vacuum line

#1 Post by Lpg08 » Mon Jan 04, 2021 9:12 am

Hi guys just wandering the purpose and importance of running a vac line to the converter?

I have seen some installs that t into the map line and other just leave it vented to atm with no issue.

Whats the go? And whats the advantages of running the line


Posts: 4071
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire

Re: Converter vacuum line

#2 Post by LPGC » Mon Jan 04, 2021 2:54 pm

You're talking about the difference between gas pressure being referenced to atmospheric pressure (no line to reducer) and gas pressure being referenced to manifold pressure (with line to reducer). The majority of systems have the line.

Most injectors work best with a pressure difference between gas feed and outlet below around (say) 1.7 bar, a commonly used pressure is 1.2 bar. If reducer gas pressure is set to 1.2 bar referenced to manifold pressure the injectors will usually see 1.2 bar pressure across them (inlet versus outlet) regardless of engine manifold pressure BUT if reducer pressure is set to 1.2 bar referenced to atmosphere the injectors will see 1.2 bar when the throttle is completely open (when manifold pressure = atmospheric pressure on a normally aspirated engine) but might see 1.9 bar at idle (when manifold pressure might be 0.3bar... 1 - 0.3 = 0.7, 1.2+0.7 = 1.9). On a turbo engine that can see positive manifold pressure the 1.2 bar referenced to atmospheric pressure would only be 0.2 bar if the inlet manifold sees 1 bar of boost pressure.

Because of this, even on systems that don't normally have reducer pressure referenced to manifold pressure on a normally aspirated engine (e.g. Prins VSI1) you'll see the reducer has often been modified to fit a vacuum/boost pressure reference to act on it's diaphragm if it's fitted on a turbo engine. And because most injectors don't like high pressure (they may fail to open at all if pressure is too high, but well before they fail to open the response of the injectors and hence fuelling accuracy will suffer) most systems have the reducer referenced to atmospheric pressure.

The petrol system on most older vehicles has petrol pressure referenced to manifold pressure. But for quite a few years now vehicle manufacturers have increasingly referenced petrol pressure to atmospheric pressure. This is a bit like the petrol system equivalent of us (LPG installers) fitting a vacuum connection (line) to the reducer or not fitting the line... So it may seem that we should connect the line on engines where petrol pressure is referenced to manifold pressure and not fit the line on engines where petrol pressure is referenced to atmospheric pressure but it's not as simple as that (due to points mentioned above about injectors not liking high pressure difference across them etc) and in practice it can still be far better to connect the line even on engines that have petrol petrol pressure referenced to atmospheric pressure. If we do connect the vacuum/boost reference line on engines with petrol pressure connected to atmospheric pressure we should set up the LPG ECU to account for the difference between petrol pressure being referenced to atmospheric pressure while LPG pressure is referenced to manifold pressure but this isn't possible with a lot of LPG system ECUs and a lot of installers haven't even considered this...

If we double pressure to a nozzle, petrol or LPG, the nozzle will flow around 1.4 times as much fuel as with the original pressure. But so far here nozzle refers solely to the nozzle not the injector as a unit. Injector units comprise the nozzle and also something that will close to shut off flow through the nozzle (when there's no electrical power to the injector) and open to allow fuel through the nozzle (when there is power to the injector). If anything effects the speed of this opening/closing it will affect how much fuel the injector flows for a given electrical pulse length, i.e. it will affect the injector's response / fuel dosage for a given pulse length.. The response of most petrol injectors isn't effected much by different fuel pressure or temperature but the response of LPG injectors can be effected to quite an extent for changing pressure and temperature (and anything that affects opening/closing times will make most percentage difference to flow for a short pulse time). If petrol pressure or LPG pressure are referenced to manifold pressure we don't have to account for manifold pressure (in terms of how it will effect injector flow) when setting up fuelling maps because fuel pressure minus manifold pressure will always be the same (*simplified point omitting other variables). But if we have a situation where, say, petrol pressure is 3.5bar referenced to atmospheric pressure while LPG pressure is 1.2 bar referenced to atmospheric pressure we still have a situation where petrol pressure referenced to manifold pressure fluctuates between 3.5bar at full throttle and 4.2 bar at idle with 0.3 bar manifold pressure while LPG pressure (reference to manifold pressure) fluctuates between 1.2 bar and 1.9bar.. which is the equivalent of petrol pressure falling by around 15% from idle to full throttle while LPG pressure falls by around 37% between idle and full load. Or if we reference reducer pressure to manifold pressure we still have a difference of 15% for petrol pressure but now a difference of 0% for gas pressure. Clear to see that there's more difference between 15% and 37% than there is difference between 15% and 0% even if we didn't have to account for LPG injector response being negatively effected in over-pressure situations. One way of accounting for the 15% range of petrol pressure in LPG calibration (if petrol pressure is referenced to atmosphere while LPG pressure is referenced to manifold pressure) is to increase LPG fuelling by a few percent for low manifold pressures using LPG ECU's that feature fuelling compensation for manifold pressure reading. The few percent has to be calculated based on petrol pressure and range of manifold pressures - the manifold can see lower pressure than idle pressure during over-run type situations, manifold pressure might go well above atmospheric pressure on a turbo/supercharged engine, petrol pressure might be anywhere between around 3 bar and around 5 bar.

As a rule of thumb if your LPG bits have a manifold pressure sensor and a reducer vacuum port you should connect them to manifold pressure. Most systems except Prins will have such sensor/port. There are few exceptions for a proper job and good results but one such exception was old advice for e.g. Valvetronic engines (still better to connect to manifold vacuum even on Valvetronic type engines if the ECU allows the compensation for petrol pressure).

*Simplified point omitting other variables - other variables include aspects like: Mixture enrichment for acceleration / Slightly less manifold vacuum due to tight engine during warm-up where the petrol ECU anticipates the higher map and increases pinj more than it would if petrol pressure were manifold pressure referenced. Regards acceleration enrichment imagine the petrol system designer wanted to factor in 10% mixture enrichment for acceleration from low load say 4ms to 8ms sudden increase in throttle position (10% on top of the already doubling of injector pulse time), so at the point the system would normally pulse injectors for 6ms it might pulse them for 6.6ms. The 6.6ms is going to be effected by manifold pressure at the time of the injection pulse during which conditions are rapidly changing, if the fuel system designer had found that manifold pressure would be increasing quite slowly at this time he might have toned down the 6.6ms to only 6.3ms, or if the manifold has low volume and pressure increases more rapidly during this tip-in the 6.6ms might be 6.8ms... For the LPG system to properly account for the amount of extra enrichment the petrol system designer wanted, if petrol pressure is atmospheric referenced while LPG pressure is manifold referenced the LPG system will have to account for manifold pressure at the time of this enrichment during the changing conditions (and have a high sampling speed of pressures etc) / Particularly if the petrol ECU makes use of a map sensor for calculating fuelling the amount of air the engine will get through (and fuel it will get through) is based not only on manifold pressure and a direct link to rpm, it is also effected indirectly by rpm because the engine's volumetric efficiency is different at different RPMs due to aspects such as cam timing and cylinder head and valve train design. There is interplay between RPM, position of VVT solenoids etc (on some vehicles), and aspects I've mentioned above such as injector response over various temperatures and pressures. Each individual factor in itself doesn't make much of a difference but if we don't address all factors properly the individual factors can add up and make for an install with issues such as widely varying fuel trims (perhaps particularly around idle) between warm-up and fully warmed up, or a system that generally seems calibrated OK but has a specific problem such as hesitation during acceleration.
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,

Post Reply