I know this thread is old but I hope to provide an answer that may assist future visitors

Before I begin answering though, I need to make a disclaimer that I am representing Hitachi Solutions which is the company behind MasterPlex ReaderFit.

Ok, now down to business. If you are not getting a good curve fit with the 5-PL model equation, there may be reasons other than the software (in this case XLfit).

First, the curve fit will only be as good as the data that you feed the software. If you do not have a nice set of points that can help the algorithm decide the best curve fit parameters, you will end up with a pretty bad curve fit.

Second, if you are using 0 standards, it will cause an error in the model equation because of mathematical limits (i.e. division by zero). In this case, I would recommend either placing in an extremely small positive value such as 0.000000001 or exclude the point all together.

Third, there could be user error in using XLfit. Unfortunately, I cannot help here but I can recommend MasterPlex ReaderFit which is designed around our core competency: Our 4PL and the 5PL curve fitting algorithms. It also includes weighting algorithms to offset one of the assumptions of the 4PL/5PL model equations which is that all standard groups have the same scatter (or variance). If one is doing an immunoassay or bioassay, we know that this is not true. In fact, we see the exact opposite and this phenomenon is know as heteroscedasticity. Try saying that 3 times fast

Please refer to my blog post for more details on the 5PL model equation or the importance of weighting with the 4PL and 5PL.

I'd also be more than happy to answer any questions you may have at aliu at miraibio dot com.

I hope this helps!

Allen Liu