The problem with the oven is simply its control system, and here the defect is obvious too, the "temperature" input has zero hysteresis and the control is "0n-off". Whoever designed this is obviously incompetent! For an oven, heat loss is quite slow in response to temperature, so a half degree of hysteresis would be much better than any mechanical thermostat and perfectly adequate, one or two degrees similarly if necessary. A system with very high sensitivity to a variable (in this case temperature) and instant full scale response feedback is basically unstable and will show all these defects as described. The best system would be one with proportional feedback with the elements controlled by a PWM IGBT or thyristor power control (pretty cheap), which would be more accurate and not pulse the mains power level!

I suspect that something like a PID controller is exactly what we have there - and it is indeed using pulse width modulation (if on a 1s cycle). Thyristor control is also bang on/off - even if you were to do it within each mains cycle (dimmer switch style) you could still be looking at very similar levels of current switching especially when it was asking for 50% power (i.e. had to switch on at the peak of each half cycle). The higher frequency might mean that it's not so noticeable on the likes of incandescent lighting, but I suspect a lot of other equipment (electronic/audio) is still going to get upset.

I agree that the design could be a lot better for single-phase application though - given the appliance almost certainly has 3 separate element, just applying the proportional control to one element and having the other two either on or off would seem to be a much gentler approach, but presumably needs a bit more thinking (and money) than just taking a 3-phase design and bolting L1-L2-L3 together, which I suspect is what we have here.

   - Andy.

The problem with the oven is simply its control system, and here the defect is obvious too, the "temperature" input has zero hysteresis and the control is "0n-off". Whoever designed this is obviously incompetent! For an oven, heat loss is quite slow in response to temperature, so a half degree of hysteresis would be much better than any mechanical thermostat and perfectly adequate, one or two degrees similarly if necessary. A system with very high sensitivity to a variable (in this case temperature) and instant full scale response feedback is basically unstable and will show all these defects as described. The best system would be one with proportional feedback with the elements controlled by a PWM IGBT or thyristor power control (pretty cheap), which would be more accurate and not pulse the mains power level!

I suspect that something like a PID controller is exactly what we have there - and it is indeed using pulse width modulation (if on a 1s cycle). Thyristor control is also bang on/off - even if you were to do it within each mains cycle (dimmer switch style) you could still be looking at very similar levels of current switching especially when it was asking for 50% power (i.e. had to switch on at the peak of each half cycle). The higher frequency might mean that it's not so noticeable on the likes of incandescent lighting, but I suspect a lot of other equipment (electronic/audio) is still going to get upset.

I agree that the design could be a lot better for single-phase application though - given the appliance almost certainly has 3 separate element, just applying the proportional control to one element and having the other two either on or off would seem to be a much gentler approach, but presumably needs a bit more thinking (and money) than just taking a 3-phase design and bolting L1-L2-L3 together, which I suspect is what we have here.

   - Andy.

The same level of current switching, but with a PWM cycle period of a few minutes to kiloseconds, rather  than  every 50 cycles would be perfectly acceptable. The problem is not the simply magnitude of the current step but the fact that it is at a rate that compares to the sensitive region for human flicker sensitivity. There is a reason for the standards I referred to above being worded the way they are !.

Now, this kit does not really comply for connection to UK mains. and as such should never be CE or UKCA marked or placed on the market.

A very nice thing to do would be to run the elements in parallel for fast heat and then in series for maintainence, but that too would require some design competences.

Mike.