Calibration problems

[DocHollywood]

Hey Guys,
Sorry if this is a novice question but I'm having some problems getting an accurate calibration curve for my GC. I'd like to detect low levels of methane with a plot U column and an FID detector with a 5890 series 2.
To calibrate I'm using two mass flow controllers. One hooked to a He line and the other hooked up to a 1% CH4/He line. I'm using a bubbler at the end of the line to measure volumetric flow rates. My problem occurs when I start getting down to low ppm levels of methane. What i do is lower the flow rate to almost a minimum for the 1% line. I'm forced to replace the mass flow controller on the other line with a needle valve to get high enough flow rates to get a proper low level standard. None to say the least these low level standards aren't giving me reproducible results when injecting them (via a 1ml sample loop) Any suggestions or better techniques?
Thanks
AD

[AICMM]

DocHollywood,

What I would suggest is a serial dilution. I have not done it but have worked with someone who has, you take the product of your first two flow controllers and T those into a second line of diluent. Set the two mass flow controllers to give you a reasonable (consistent) dilution then mix that flow with a third flow.

Make sure you are sampling at the same pressure every time as well.

Best regards.

[DocHollywood]

Great Thanks,
AD