Amazon + Airbox + Blos

[Bristol603]

My 1966 Volvo Amazon 122S was converted to LPG in Sept 2012. The setup was a simple open loop system (45+50 litre tanks, OMVL R90e vapouriser, power valve and twin mixers for the two HS6 SU carburettors). A problem with fitting LPG to my car was the lack of vacuum at the mixers to draw the LPG from the vapouriser- each SU is only feeding half of a 1.8 litre engine. The setup used had a high idle setting on the vapouriser to help push through some LPG at low RPM and an extra obstruction (i.e. bolt on a stick) held in the mixer opening to try to create more vacuum. This worked well enough in daily use, but measuring with a lambda sensor showed it was running lean above 3000rpm.

One suggestion to help was to further restrict the mixer openings so as to generate more vacuum. The HS6 SU has a nominal size of 1.75 inch – I measured the inlet as 41mm in diameter. The mixer fitted was 34mm in diameter. I added a 21mm diameter obstruction in the centre of the mixer giving a remaining cross-section equivalent to a mixer with a diameter of 26.6mm. This still didn’t provide enough vacuum to stop the mixture leaning out at high revs. Obstructing the inlet also hinders maximum airflow and performance. I wanted another solution.

Some time on the internet showed up a device called a Blos LPG carburettor – they seem widely used on Land Rovers. The Blos is similar in operating principle to an SU, but works with LPG rather than petrol and relies on the existing petrol carburettor butterflies to regulate airflow rather than having its own butterfly and throttle linkage. In place of the piston in an SU, it has a sliding plate. The venturi in the Blos is about 50mm in diameter. The cost was £100 mail order from Poland – by comparison two mixers suitable for a pair of HS6s will cost about £50.

The problem was then how to connect a single Blos LPG carburettor to the twin SU carburettors. The solution I selected was a custom airbox. This was made from 3mm sheet aluminium. The Blos has a very wide opening at 70mm and so a 70mm inlet flange and some 70mm silicone hose were used to connect the Blos to the front of the airbox.

Although I had never had any backfires with my setup, I was conscious that there would now be an airbox full of gas and this could make any backfire much worse. The airbox itself was quite robust, but I still needed somewhere for the expanding gas from any backfire to dissipate. I decide to fit a blow off flap (essentially a 50mm diameter one way valve) in the side of the airbox facing down through the engine bay to the road. I also spring loaded the lid of the airbox into place rather than bolting this down – vacuum generated in use sucks it in to place and should a backfire occur (and the blow off valve be insufficient) , then the lid can lift and release the pressure.

The bottom and lid of the airbox were edged with closed cell draught excluder foam to make the airbox leak free. When assembled, the holes for the carburettors were taped over and the airbox filled with water to test for leaks. A little adjustment with a hammer resulted in a watertight, and hence airtight, unit.

Airfiltering is provided by some stainless steel gauze over the inlet to the Blos and some Ramair foam filter media held by shaped mesh inside the airbox. The minimum clearance when fitted in my car is about 20mm from the end of the brake master cylinder.

My car started first turn of the key using the approximate settings recommended for the Blos. The fine tuning is easy if you have a lambda probe. You rev the engine to 3500rpm and turn the mixture wheel until you get a rapidly varying lambda signal averaging 0.5V. You then idle the engine and turn the idle wheel until you get a rapidly varying lambda signal averaging 0.5V. Both settings are then rechecked in case one has slightly affected the other.

The car drives and idles very well using the Blos and custom airbox. Throttle response is quick, the 2000-3000rpm torque good and the mixture does not lean out at high rpm. I am happy with how the change worked out.
I am told by those more knowledgeable than I that an airbox with a ring type mixer of around 34mm diameter may work as well or better than the Blos. The airbox forces the air drawn in through both carburettors to flow though one mixer and so increases the vacuum generated. It would also allow a more efficient ring type mixer to be used in the inlet to the airbox rather than individual slim plate mixers sandwiched between the air filters and the carburettors.

If I was making the airbox again, then I would taper it in depth moving toward the bulkhead to reduce the airbox volume and increase the clearance from the hydraulic master cylinders.

My simple open loop system returned 26mpg measure over 8000 miles. I do not yet have an accurate figure for the new setup.

[940_Turbo]

Are you not worried about the rear cylinders running lean? I'm not convinced about the asymmetrical arrangement at all.

[Bristol603]

I hadn't been worried up to this point!

The LPG and air mix at the Blos and so I don't see how the LPG/air mixture ratio could change for 3+4 vs 1+2 as it is already mixed by the time it gets into the airbox.

The petrol and air mix at the SU and I don't see how the airbox would change the petrol/air mixture ratio as this is controlled by the air flowing through each SU.

With respect to charge mass, I was working on the theory that the cross-section of the airbox (approx 110mm*110mm) is sufficently large compared to the volume of air and fuel being drawn through each SU that there is effectively no pressure differential front-to-back inside the airbox. If there were any pressure difference, then I "think" this would mean 3+4 would get a lower charge mass of air+fuel, but wouldn't change the mixture ratio.

How does the original Volvo airbox for a twin SU setup with the cold air intake at the front address the issue? I have seen a drawing of this in the Amazon parts book. The air inlet is at the front and each SU draws from the airbox, i.e. 3+4 are further away from the inlet than 1+2. Does the Volvo airbox cause leaning running on 3+4?

If I have done something stupid, then I would really like to know before I damage anything.

Thanks.

[940_Turbo]

The twin airbox for the SUs doesn't need to address the issue. It's supplying an effectively limitless quantity of air to two carbs. Carburettor number one feeds cylinders 1 and 2 and the second carb feeds the rearmost cylinders. The twoi carbs are linked with a balance pipe.

Your Blos carb doesn't know which cylinder is firing; that airbox is fed with a constant supply of air/fuel mixture.

Think about the firing order, #1 fires nearest to the the fuel supply, then #3 fires followed by #4 and finally #2.

Cylinder #1 has a far easier route to draw in fuel than the rearmost cylinder. As far as I can see because of the unbalanced position of the carb #1 must run richer than #4.

That is why a single carb is positioned centrally to the engine, not off to one side. I can't think of any exceptions to that rule.

[Bristol603]

Thanks for taking the time to reply.

I'm not sure I understand how carb 1 can get a richer LPG/air mixture than carb 2 as the LPG and air are already mixed together as they enter the airbox. How would mixed LPG/air exiting the Blos know that it would end up going through carb 1 or carb 2 and somehow change its mixture ratio? I am sure you know loads more than me, but I don't understand how the mixture ratio can change after the LPG and air have already been mixed.

I agree that single carbs are placed centrally, but thought this was more due to balancing the flow paths downstream of the carb rather than anything upstream of the carb.

Any enlightenment you can give me would help.

[Bristol603]

PS - The Blos doesn't have a throttle valve. It supplies an unlimited quantity of LPG/air mixture into the airbox as demanded by the SUs. The flow to the cylinders is still controlled by the throttle valves of the SUs. The volume of the airbox is about 4000cc - it is a plenum chamber filled with LPG/air mixture.

[940_Turbo]

Maybe I'm wrong; however my limited understanding is that a finite amount of fuel/air can enter your plenum. Your Blos carb is in essence similar to the SU in that it has a variable venturi.

Each cylinder takes its share, therefore there will be a pressure gradient from front to rear. There are also pressure waves, the mixture will accelerate as it is drawn into the narrow inlet port and as the valve snaps shuts it will bounce back. That is the reason cars use tuned length inlet manifolds.

I don't know whether there is a real world effect of positioning the Blos as you have. It just looks very odd to me hence I questioned it.

[tom-madbiker]

in order to fit my blos to my amazon i have constructed up a new manifold because i was not happy with what i had although i am running a different engine (b6304) and i am going to run only gas for the moment i used a throttle body and a 90 deg pipe to feed the manifold still to try the setup but looking ok i will take a photo different engine but the principals the same

[classicswede]

As the gas is dry the mixture will be even across all cylinders, better than what you see on petrol infact as the air speed is constant for all.

Petrol however would show variation IF you had condensation. A lot of forced induction systems run a carb at one end of a manifold and condensation is what tends to cause variation in those setups. The variation in those cases is usualy not enough to cause a problem.