LPG Discussion Forum

Q: In a single-point set-up when both are possible which is optimum?

Do we plonk the mixer, engine side of the carb, or air-filter side? I've seen both. Example: there's purpose made cooker-rings for the SU carb, and I think I've seen twin-choke Weber variants too? Because they choke-off WOT I've always avoided such things. I can see advantages to both, if there's the option, which is better and why?

Reason I ask is that next I'll be fitting a Holley Sniper Autolite 1100

https://www.holley.com/products/fuel_sy ... ts/550-552

We can't fit a mixer on the vacuum side of a throttle plate for the obvious reason that the vacuum would cause the reducer to dump loads of gas in through the mixer and make the mixture vastly too rich.

We could fit a mixer between a carb and throttle body if the setup has separate carb and throttle body like on some of the Holley etc setups... But the carb would probably mess with airflow through the mixer and potentially cause problems for running on LPG - A similar effect to the reverse... Mounting a mixer on top of a carb too close to the carb can airflow through the carb and mess up running on petrol, in which case a spacer between the mixer and carb underneath it can help prevent problems.

Is the Sniper going on the Landrover? The sniper is a kind of fuel injection system and uses a sensor to detect manifold pressure etc, it doesn't have to rely on things like the venturi effect to suck fuel out of jets like a carb does so will be less sensitive to mounting a mixer close on top of it than a carb would (and in any case is mappable so you should be able to get good results anyway). I would fit an LPG mixer on top of the Sniper if I was fitting a Sniper.

LPGC wrote: Tue Jul 23, 2024 6:40 pm We can't fit a mixer on the vacuum side of a throttle plate for the obvious reason that the vacuum would cause the reducer to dump loads of gas in through the mixer and make the mixture vastly too rich.

Of course - thinking more clearly it was a stupid Q. Those SU mixers are pre-carb.

Yes, currently with an HS6 I'd like to put the Sniper on my LR. Not this week but sooner or later. I won't be the first to do it. I've heard good things for their single choke offering. And again, I take you point. What with the Sniper, not vacuum dependent, provided the mixer can be made to work, the EFi system can be set to 'walk' round it.

Thank you.

Been thinking about making electronics that could drive a sequential fuel injection system on almost any engine. Normally sequential injection (firing individual injectors just as inlet valves open on respective cylinders) requires a cam sensor so the system knows where each cylinder is in it's 720degree 4 stroke cycle but we can get the same information with easily good enough accuracy by monitoring when individual spark plugs fire. We can get info about when individual spark plugs fire easily by using an inductive pickup on spark plug wires (totally accurate and the old skool method by which timing lights worked). Or on coil on plug engines we don't need to use inductive pickups, just use direct connections to the coil packs (though any engine with coil packs is going to have fuel injection anyway defeating the object).

If you have (say) a 4 cylinder engine you''d have 4 inductive pickups one for each spark plug, 4 outputs. The start of an output pulse would be triggered by reading an input pulse from it's relative spark plug wire, outputs would have variable length pulse width with pulse width linked to either manifold pressure (map) or airflow (maf) divided by rpm. The input from one cylinder would cause output on another cylinder because we don't want to start injecting fuel on a cylinder that has just begun its firing stroke, instead we want to be starting injection on a cylinder that is just about to start its intake stroke, so on this example 4 cylinder engine assuming it has typical 1342 firing order when it reads an ignition pulse on cylinder 1 that's when cylinder 4 will be beginning its induction stroke so we'd set it so input on cylinder 1 would cause output on cylinder 4.. Though actually the gadget wouldn't even need to address that because the installer could/would just wire it so output from cylinder 1 input went to cylinder 4.

This gadget itself doesn't need to be much more sophisticated than that because the variable length output pulses could serve as input to an off the shelf ECU such as an LPG ECU which provides the mapping capabilities (so the gadget itself doesn't even need to be adjustable/mappable). An LPG ECU could drive LPG injectors or petrol injectors and the overall system would still offer the essential full mapping capability to adjust fuelling for all the various loads and rpm combinations.

This doesn't address the fueling components install (injectors, fuel rails, etc) but you could modify an existing inlet manifold to fit LPG or petrol injectors (just be careful of carb engine inlet manifolds coolant channels), would probably work out cheaper than something like the Sniper and offers superior sequential injection instead of throttle body injection.

LPGC wrote: Thu Jul 25, 2024 7:21 pm

If you have (say) a 4 cylinder engine you''d have 4 inductive pickups one for each spark plug, 4 outputs.

So long as it isn't double ended coils with so-called 'wasted spark'.

I shudder to think of the R&D cost for such things. The idea is the easy bit, who knows what Holley's Sniper cost to get to prototype. Reckon, wrong side of half million, no worries. Before production, likely over one big one all told?

CNG wrote: Thu Aug 08, 2024 9:16 am I shudder to think of the R&D cost for such things.

I'd reckon on about £50 in bits from RS if it was done as an 'out of interest to see if it would work' project. If done by a company that is counting the costs of time then yes, it could get expensive, but not if it was done by an individual not taking into account time.

Many years ago me and a colleague built what we reckon was probably the worlds most expensive, programmable Christmas lights. Long before the likes of the Raspberry Pi, this was all discrete components and could drive 16 individual 24V coloured lights and make them create patterns by a sequence of switching them on and off. It took two of us about 5 months to get two working prototypes built and the total cost must have been all of £30 a unit in components but time wasn't factored in. We just had 2 houses that had a light show that was fascinating in those days but commonplace now using off the shelf flashing LED arrays now.

Like Gilbert I was thinking it'd only need cost around £50 for the bits. Remember it's not the total setup, it's just something that will drive a slave injection system. The slave system would have an inexpensive off-the-shelf ECU handling the mapping and injector driving (many off the shelf LPG electronics would be ideal), the fitter would have to source and fit none electronic components such as the injectors themselves but particularly for an LPG vapour injection install that wouldn't be difficult or expensive at all.. it would only be the same parts we'd fit in the same way as for 'normal' sequential LPG conversions. For a sequential petrol injection setup you'd need the petrol injectors, electric fuel pump, some way of installing the petrol injectors on the inlet manifold to get the spray pattern into the airflow while preventing vacuum leaks and fuel leaks. If there is a fuel injected version of the same (carb) engine it might be possible to make this simple by just using the fuel injected version inlet manifold and injectors setup.