ROTAX 912 low fuel pressure at take off

Not having seen the fuel system for a Sonex with wing tanks, I just assumed the wing tanks fed into the header tank.  However, if I understand you correctly, your wing tanks and header tank are selected independently (one or the other, but not both), and only the wing tanks use a booster pump.   

Does the low fuel pressure problem ever happen when you are using the header tank? If not, then it may be that the wing tank boost pump is insufficient as Roger suggests. It may also be that the flow through rate is too low with the boost pump turned off. If you have not already done so, you could simulate the climb-out at a safe altitude with the header tank selected.  You could also disconnect the fuel supply line at the mechanical fuel pump, and check the flow rate from the wing tanks, through the boost pump, all the way to the mechanical pump inlet. To do this with the boost pump off, you may need to lower the end of the fuel line and create a siphon. It would be interesting to see if the flow, with both the pump on and off, is at least 37 L/hour (rotax fuel pump flow rating).   

Jeff B‘s hint regarding the bypass of the boost pump should be taken into account, I believe. The flow of/through the boost pump was described as „not wonderful“. This may be an indication for some malfunction of the boost pump.  Even if it is not blocked completely it may restrict the capabilities of the mechanical pump and cause poor supply of fuel. 
If there was a bypass to the (probably malfunctioning or poorly performing) boost pump the flow of fuel would go around (not through) the boost pump and the mechanical pump would be able to do its job with no restriction. 
The bypass solution suggested by Rotax costs a short piece of tube, two T-joiners and the little bypass valve that hinders the fuel to go through the boost pump in circles (from its outlet to its inlet). At the same time it covers boost pump blockages. This setup is recommended. If you Sonex doesn‘t have it yet you should go for it anyway. 
Measuring the performance of the two individual pumps and their joint inline throughput completes the analysis, I recon.

Hi Peter KI,

I agree a boost pump fuel bypass or even better a separate boost fuel circuit, is highly desirable and I may fit one or the other "down the track".

I do not agree with your diagnosis - my main/mechanical pump delivers 4.5-5 psi during pre-flight testing. Add the boost pump ON and it goes up by a further 2-3 psi. 

True- the presence of the, in line, boost pump may cause a slight flow restriction and this may be a contributing factor to low fuel flow/pressure during TO phase, however a 7.6L/hr fuel return line seems, in my mind,  to be a larger factor.

The anemic gravity flow (which as tested by independent assessor, exceeds minimum fuel flow) from the wing tanks(s) through the boost pump is because they are all pretty much at the same drain level - there is no blockage.

The boost pump ON,  flow is not much better because the boost pump is primarily a pressure pump, that has poor performance when there is an open fuel circuit - squeeze/constrict the end of the open fuel hose & you get a jet/spray.

I am still waiting for my Rotax dealership to get back to me with the ID of their #35 restrictor jet.

Hi Sean,

7.6l/hr seems very high - I fully agree that it needs „treatment“ and you are at it already.

My focus was more on the influence of a potentially defect boost pump with no bypass being used in-line with the mechanical pump.

Please let us know the details of the jet once you know.

Rotax supplied fuel restrictor jet delved today PN 963820 PILOT JET 35.

ID is .53mm - same as the one already in my fuel return.

Will have to find a smaller ID jet from an alternative supplier.