We have, however, found a way of avoiding short firing. There are also ways of masking the problem.
Avoiding short firing with Pump Delay Time
There is a Pump Delay Time setting on screen C21 of the VRC430f. By default it is set to 15 minutes. The effect of changing it to OFF is shown here:
(the parameters are the same as for the graphs shown on the page called "The problem")
This change therefore seems to avoid short firing altogether, and (in the absence of conditions leading to micro firing) to achieve very long burn times. Since the boiler is now on for more of the time, this also allows the heating curve to be set to a lower value.
We do not know what the pump delay time is for. We can guess, but if you actually know, we'd be pleased to hear from you. Note that this change does not, so far as we know, affect micro firing behaviour.
Masking the micro firing problem
Since micro firing appears only when the target flow temperature is in the low 40s or below, it is possible to avoid it by setting a higher minimum flow temperature. It can be altered on screen C9 of the VRC430f. People have variously reported success by changing it to 50C or 55C or even 60C. Then the boiler can modulate successfully to this temperature without turning off.
However, this is not really a solution at all - it is merely hiding the problem, and moreover it is losing some or all of the advantage of weather compensation. Under ordinary circumstances, the flow temperature with weather compensation will nearly always be below 50C. By forcing it to stay at this temperature or above, the system cannot work as intended, since it now cannot balance the heat loss from the building with the heat input. In other words, you might just as well have stayed with on/off controls, since that's effectively what you're moving towards by making this change.
Return sensing
One user has reported that micro firing disappeared upon changing his boiler from flow sensing to return sensing (i.e. changing parameter d.17 from 0 to 1). Both authors of this blog have tried this change without improvement - see post here. The difference may be system-specific, or may be due to the fact that we use a VRC430f controller whereas the beneficial change was seen with a VRT392. However, it's just possible that it may yet help in our systems - previous experiments were tried before we abolished short firing by changing Pump Delay Time as described above. We'll therefore try return sensing again when we have a chance and report back.
d.2 boiler parameter
This is another change that others have reported to help in respect of micro firing. It affects the period that the boiler waits before trying to reignite. However, in our hands it affects only short firing (and 'normal' cycling), but does not help in respect of micro firing - see here.
