Troubleshooting Extruders, Part 6
In my last column, I began to review how the controller varies power to the heater. I'll conclude this series with power feedback, heat and cool cycle times, balancing heat and cool, and controlling cooling zones.
Suppose your zone is running at 25 percent power and the temperature is right on setpoint. Then, suppose line voltage falls some 20 percent. This results in a heater power drop of 36 percent because of the square law dependence of power on voltage. Sooner or later, the temperature falls.
After a time, the thermocouple and controller would sense this fall and increase the “on” time of the contactor just enough to bring the temperature back to setpoint. Meanwhile, the material is running a bit cooler than optimum and may show some imperfection in the product.
A smart controller would watch line voltage continuously and increase or decrease contactor percent “on” time to compensate right away. In this way, the zone need never suffer a temperature disturbance caused by a line voltage change. Remember that the feature, power feedback, is only applicable to electric heaters. Disable it if you have some other heating medium such as water, steam or heat transfer oil.
In process-control terminology, the same idea -- heading off the effect of a disturbance as soon as it happens -- is called feedforward. For example, you might vary the base heater power on a conveyor furnace automatically in proportion to material throughput and leave the controller to do only fine trimming and correction of other disturbances.
Choice of Heat Cycle-TimeThis is a controller parameter with a range adjustable typically between 0.1 and 100 sec. A 0.1 sec time (10 operations per second) would soon destroy a magnetic contactor, so in this case, you would select typically 20 sec. On most extruder zones, there is so much metal mass that the temperature would not rise and fall noticeably in sympathy with these slow “on” and “off” pulses.
On the other hand, consider a fast-responding application like infrared radiant heating of moving sheet. You would need a cycle-time of say 0.2 sec to avoid the alternately under- and overheated sections of sheet that slow cycling would give. A 10-sec cycle would show large temperature swings. On any fast control loop, be on the alert to recognize cyclically under- and overheated sections as too long of a cycle-time -- not control loop instability.
If you are using tungsten lamp heaters, even on a slow loop, you would not use magnetic contactors. Every switch-on would give a massive overcurrent. This kind of heater would call for silicon-controlled rectifiers (SCRs) with phase-angle control in the manner of a lamp dimmer, and possibly a current-limit feature.
Choice of Cool Cycle-TimeIf you are switching a solenoid valve or blower motor with a magnetic contactor, a cycle time of 10 or 20 sec would be suitable because of the mass and slow response of a typical cooled zone.
Some machines have blowers, switched by fast solid-state contactors (e.g., 0.1 sec cycle-time). Instead of stopping and starting every cycle, the blower, because of its inertia, will assume some intermediate speed at which the controller is satisfied that the temperature is correct.
Balancing Heat and Cool DeliveryIt is rare to find an extruder zone whose cooling capacity is equal to its heating capacity. You can see this if you switch on the heater band and record the rate of temperature increase; then switch on full cooling and record the rate of fall. If the rate of fall is say five times the climb rate, then the cooling capacity is five times the heater capacity.
In this example, a temperature rise would bring a dose of cooling five times as strong as the heater would deliver for the same undertemperature, and you would see a severe drop of temperature, then a recovery and continual repetition of this temperature cycling. The solution? Run in heat-only mode and tune the control loop; that is, find the PID settings that give stable control.
Now suppose the cooling proportional band is five times that for the heat mode. This translates to setting a cool gain of 0.2. The parameter for this is sometimes called relative-cool. The mnemonic (abbreviation) in the controller parameter list depends on the brand of controller.
Heating capacity is easily defined (the kW rating is stamped on the heater band).However, cooling device capacity is not easy to define or to keep constant.
Water CoolingCooling capacity with water depends on inlet water temperature, working temperature of the zone, flow rate and heat transfer of the cooling jacket.
Inlet temperature can change during the day, dependent on other loads on the cooling water system. Maximum flow-rate is often changed by a throttling valve according to the judgment or experimentation of the operator. Unless you know the machine design, the quickest way is, after tuning in the heating mode, find a setting for cool gain setting that gives temperature stability when that zone is running in the cooling mode.
Another complication with water-cooling comes if the zone is running well above 212oF (100oC). Usually the first few pulses of water will flash off into steam, giving a greatly increased cooling capacity due to the latent heat of evaporation. When the zone settles down, less or even no evaporation is a possibility and the cooling is less severe.
To handle evaporative cooling you would choose the water cool mode from the controller parameter list. This technique delivers much shortened pulses of water for the first few percent of the cooling range, when the water is likely to be flashing off into steam. This compensates for the transition out of the initial strong evaporative cooling.
Fan CoolingThis is much gentler than water cooling and not so immediate or decisive because of the long heat transfer path through the finned aluminum cooler and barrel. Again, capacity is hard to predict but you can get a rough idea if you can feel the outgoing air from that zone with your hand and compare it with how a 1,500 W hair drier feels.
With fan cooling, a cool gain setting of 3 upwards would be typical, and delivery of pulses to the blower would be linear. In other words, the “on” time would increase proportionally with percentage cool demand determined by the controller.