Step7 - pH Control with FB41

I've got to realise a ph-Control system for my current project and I intended to use a system I implemented successfully on an S5 system about 15 years ago in a waste-water treatment plant, where I simply ramped up the output (in that case a VFD controlled feed-screw, in this case a variable-stroke pump) and when the SP and MV differed by less than a certain value (obtained by trial and error) stopping the ramp and re-starting from zero. By using two ramps, effectively coarse (when SP and MV were wide apart) and fine, I was able to get things so that the control was well within the required limits.

Unfortunately, for various reasons, the customer would like me to use FB41 for the basic control, which wouldn't be a problem except for the fact that I can't fathom out how to (effectively) re-start the Integral action timer.

Anybody got any ideas?

My first idea is to have the customer sign off on responsibility for making his system work, and to make sure you are being paid an hourly rate.

pH control is very non-linear. PID is not good for non-linear systems.

Your original logic is much better for the application.

Roy

I agree with Tom about not mixing the letters P, I, D, p, and H. They don't spell anything worthwhile.

But if you're forced in to attempting this you can access a number of the PID attributes in the FB41 IDB. The Siemens model for PID calculates the P component, the I component, and the D component separately and then sums them all, subject to the P_SEL, I_SEL and D_SEL switches selecting them. For the I component specifically you can control TI (the integral time period), INT_HOLD (freeze the integral calculation and hold the output value), I_ITL_ON (re-initialize the integral calculation), I_ITL_VAL (what to use as the initial value)

I think you could effect what you need by holding the action, re-initializing with a value of zero, and then unholding. There's loads to explore in there!

regards

Ken

I_ITL_ON (re-initialize the integral calculation

Click to expand...

That's what I was looking for! There are so many blooming parameters in FB41 that I couldn't see the wood for the tree's ( or in this case, it's more like I couldn't see the tree for the wood!).

We're still discussing the problem, but I think I'm beginning to win the customer over to my way of thinking. I'm pretty sure I can "bend" FB41 to do what I want, but since one of the reasons that he wants to use FB41 is so that his commissioning people can use their standard interface to tune the loop. I've been pointing out to him that we're not going to help them much when this one loop responds completely differently to the way they expect. Far better to use a different method of control and explain to the commissioning people in detail how it works, what parameters they can vary and what exactly they do.

Problem solved!

The customer has now agreed to do it my way, which makes life a lot easier!