Hello All.
I have recently installed an On-column Injection port in my HP5890 GC, Where a 30m, 0.53mm column is connected.
The detector is FID.
Can someone recommend starting parameters for this configuration. Most method I used before were made in split 1:100 ratio.
Injecting Normal samples in the new injector are coming up with very high and broad signals.
Should I install a column gap?
What flows should I start with ?
Any help is very much appreciated.

If you are injecting the same concentration that you did before, you are more than likely overloading the column. You are now doing a splitless injection, where before you split 100:1, so you are putting 100 x more compound on column. Calculate the mass of compound on column, then compare that with the column capacity to determine if you are overloading the column.

I would try diluting one of the samples with the broad peaks by 50 or 100:1, then injecting to see what the chromatography looks like before changing any other parameters.

Just out of curiosity, why are you going to on-column injection when you were splitting 100:1? It doesn't sound like sensitivity should have been an issue.

The main problem with the former injection was a thermal instabillity problem. the material can not be injected by normal evaporation injection.

Thermal instability is a good reason to try the on column technique. Try diluting the samples, using a starting injection port temperature close to the boiling point of your solvent, preferably a few degrees lower, and try using the same flow and temperature ramps previously used. A retention gap may help to minimize band broadening, but I have had good results with or without one. One thing to be aware of is that there may be a temperature gradient in the injection port so that the column at the point of injection is either hotter or colder than the temperature setpoint. A longer equilibration time or a temperature setpoint adjustment may be necessary for best performance. Different manufacturers inlet perform differently, so at times you can't take a method developed on one instrument and run it on another without some modifications.

My feeling is that your samples are too concentrated and the column is overloaded, so dilution of the sample is probably the simplest way to start developing methods. If you can inject a smaller volume of the current concentration, 0.05 to 0.10 ul using a 1 ul syringe that should also help if column overload is a problem.

Good luck.

zzzz