A Modern PID Controller for an Old Electric House Furnace

So instead of simply responding by full on or off you want it to run proportionally to the difference?

If so you will need considerably more advanced control system to vary the heating element power levels plus fan speed.

If so what exactly are you thinking you would gain from this type of control system?

PID or fuzzy logic controllers can interface to VFDs (for fans) or solid-state heat controllers (for heating elements). Some VFDs and SSHCs can accept 4-20mA devices, or even thermocouples or RTDs, directly. The trick will be to find the simplest combination.

http://www.americanstandardair.com/products/add-on-components/zoning.html

Looks like CaptiveAir makes what you're looking for, but it takes more than a fancy thermostat to accomplish your goal. You'll also need some SSRs (Solid State Relays) that interface with thermostat system.The thermostat has a proportional output which tells the SSR how long to stay on in 1 second increments.

You might also look at the Nest system, it's much smarter than anything else out there. It learns your habits, the response of your heating system and the thermal dynamics of your house, then integrates that info with real time weather forecasts (also known as "Outdoor Reset" in the boiler control industry). It knows if the temperature is going to rise or fall into the future and adjusts the on-time accordingly to minimize overshooting the setpoint. Plus it's web-enabled so you can know the inside temperature and adjust it from your Smartphone. I'll bet it's a lot cheaper than any of the other systems.

On-Off controllers are fine for many temperature control applications. The temperature will usually oscillate a degree or two above and below the set-point.

Proportional only feedback control will get you very close to the set-point, but the response time may not be optimal and there may be some overshoot.

PID (Proportional-Integral-Derivative feedback) controllers can be very useful when specific process requirements are needed.
PID controllers, when properly tuned, will get you to the exact set-point in the quickest time possible with no overshoot.

If this is a learning experiment to see how PID controllers work, go for it.

If this is an effort to improve the efficiency of home "electric" heating units, IMO you are wasting time and money. But it's your time and your money!

I've used this source for temperature control products and accessories for over 2 decades and have never had any problems.

http://www.omega.com/prodinfo/temperaturecontrollers.html

The controller, a suitable temperature probe, and a solid state relay (to switch the higher current load) will be required.

The electric heating device and its fan may require modification to work with an external controller (one not installed/approved by the manufacturer, UL, NFPA, etc.). WARNING! Lethal voltages and currents are involved and there is also a potential fire hazard. You take full responsibility and liability if you make unauthorized modifications.

Most thermostat manufacturers produce a thermostat for electric base board heaters. Designed to maintain temperature .5 degrees plus or minus set point. The output of these is only designed for 3500 watts. But it could be used to drive a solid state contactor. Will need to provide 120 volts to the thermostat. I don't know that changing blower speed would be recommended. A certain amount of back pressure in the duct work is necessary to distribute heat air at all the registers. Slowing it down and you may find some rooms cold.

what do you mean by "all the way on or off"??

I'm still waiting on your response to my question. in the meantime ..if you're just trying to get a tighter control band with fewer temp swings, this is a piece of cake.

all you need is a digital t-stat that works with a remote sensor. these are very available by several manufacturers.put the actual t-stat anywhere you like and program your desired temp, and times. you can use a hardwired sensor or wireless. place the sensor near to the "return air" grill....usually 5 feet above the floor but you get to pick the exact sensing spot. as long as the blower is moving air the sensor is constantly in communication with the stat and you'll hold temp very close to your set point, the blower runs all the time, the heat or cool cycle only when the stat calls them to operate.

I want to thank everyone for their input.

My desire is to make the system a little quieter by stopping the on / off cycling.

I was hoping that given the advanced control methods that have been developed over the past few years that perhaps there were some after market controllers of this type available specifically designed for the most common conventional electric house furnaces.

I have followed the links provided but don't see a product that would accomplish my goal of variable ramping of both the blower and coil in response to the difference between the thermostat target setting and actual temperature at the thermostat.

I was looking for existing product(s) where the thermostat would send a signal to the power controller that would adjust the speed of the blower and power supplied to the heating coil in response to the difference between the set temperature and actual temperature.

Its not something I am capable of building for my own use. I need an existing product.

The ideal would be a thermostat that signals a control unit I install next to my furnace. I take the power supply line and connect it to the power input on the power controller, connect the power supply line to the blower to the controller, and connect the power supply line to the coil to the controller.

The thermostat signal would vary in proportion to the difference between target and actual temperature. The power controller would then apply variable power to the blower and coil based on that signal. Perhaps applying full power if the set temperature is two degrees above the actual temperature and ramping down to zero power when the set temperature is at or below the actual temperature.

you're describing a Ferrari control set up for your "old VW furnace", sure it could be done but thats a lot of money

Building on what others have said...Unless you have a way to modulate the airflow into each room individually, basing the output of the elements and the blower on a single control point will most likely only control one room accurately, the others will be at the mercy of your single controller.

It would far less costly to put in baseboard elements with individual thermostats in each room. If the expansion/contraction noise is the issue get a slow-ramp starter for the furnace, or use a multi-stage thermostat to bring the elements up slower combined with a two speed blower (maybe your blower motor has that capability already). You can get four steps of heat that way.

with a modern furnace you can do this type of a system but it Will not retro into an old furnace, but yes, you can control several zones and have a blower that "ramps" accordong to anticipation of load ploadhttp://www.commercial.carrier.com/commercial/hvac/product_description/0,,CLI1_DIV12_ETI434_PRD1141,00.html

Thanks for all your input.

Looks like I'm stuck with my VW but I am going to look into RAM's suggestion.

Thanks again everybody !!!

Hi Gavilan,

I looked at some of the links Fredski sent from Carrier ( I didn't know anyone made any residential stuff like that) I have a system from JCI that does the same thing in my current project -- but the controllers are a couple thousand each, each vav about the same, then proprietary stats... VFD on ahus about 4k----all runs on proprietary software, Not for home use, But one of those Carrier systems ZONECC3Z in Fredski's link could do it for you -- it would not be a walk but it is do able, if you are a fairly advanced electrically and mechanically inclined person.

What I used to do when I had forced air heat was set the fan to the low, on all the time run setting, and then let the stat do its thing-- then you didn't get woke up from the on and off all night long.

But I wouldn't ! Go for a new high efficiency heat pump with a similar system and you will save a lot of money on power and you can get all that functionality.

Best of luck,

Expat07

One might wonder why it isn't done that way already.

Thanks for the "Slack."
Well vetted doers sometimes miss the obvious? Can imagination be trained out of intellect, like a work horse being blindered to view only the path in front of its nose?
The old mechanics adage, "You can always tell an engineer, but you can't tell them much?"

https://www.facebook.com/mark.j.carter.2025

I know, I know, Facebook is only for the ignorant masses, but it allows for comments longer than a tweet and may provide a vehicle to put things into the “Public Domain” where the most efficient rule.
Although envisioned as a proportional control method used in power averaged transportation processes is there any reason this proportional control method would not work for a electric furnace controller as well as AC control?
I think you engineers call this Variable Frequency Control, but I'm not an Engineer, obviously.
Control the power applied to both fan motor and heater coil by varying circuit admittance? (1/z)
From a blurb on Electro-dynamic power processes as applied to transport processes:
Reactive Control optimizes the physics of series resonance as applied to motor/generator power control. It is applicable to many other proportional control processes.
In this proposed hybrid electro-dynamic process, power provided by the prime mover is converted to electrical energy via a crank integrated alternator, and depending on application, the output AC is first fed to a high ratio voltage step up, a full wave rectifier, and then to the power and storage circuit.
Infinitely Variable Torque Control:
Torque is controlled by varying circuit admittance.
Torque demand on the prime mover is controlled by varying the admittance of the Charging Circuit. (1/z where z is circuit impedance) Torque demand on the prime mover is proportional to Charging Circuit Admittance (1/z). Maximum torque demand is greatest when capacitive and inductive values are at resonance with the alternator frequency. The alternator frequency is defined by the rotor speed and number of magnetic poles in the alternator.
The Charging Circuit admittance can be either manually controlled by the operator via a variable control capacitance, inductance, or both; or by a process control algorithm. When combined with temporary storage this eliminates real-time demand on the prime mover and allows the operator or control algorithm to average power input throughout the entire transport cycle regardless of terrain or other demand variables, within limits. Peak tractive demand is met by capacitated energy and can far exceed the peak power output of the prime mover for short periods of time.
For a human powered process this allows the operator to choose when to input power regardless of real time demand.
For Internal Combustion Powered Processes this means optimizing prime mover mass and volume fractions. This process proposes a means to increase energy throughput efficiency, and optimize performance in land based vehicles.
This proportional control method eliminates all mechanical linkage between the crank or shaft and drive wheel(s) and allows for infinitely variable control of power and braking within the limits of wheel to road adhesion.
Input Power is fed into the system independent of real time demand and Propulsion Power is drawn off the storage capacitor based on real time demand. Regenerated brake energy is recovered and returned to storage. The same proportional control methodology is used in initial input power, propulsion and regenerative braking.
Below is the familiar equation for RLC series resonance. As stated above; maximum torque demand on the Permanent Magnet Alternator (PMA) occurs when the PMA output frequency is at resonance with the charging circuit. Maximum propulsion or regenerative braking force is developed when the wheel integrated Permanent Magnet Motor/Generator (PMMG) output frequency is at resonance with the propulsion or regenerative braking circuit. f=1/(2π√LC) in HZ or ω=1/√LC. This is controlled by the operator or algorithm.
In the theory circuit as shown, the system charging circuit, power circuit, and regeneration circuit are so interconnected as to allow the charging circuit to function independently of propulsion or regenerative braking.
Prime mover input energy can be continuously or intermittently input regardless of the propulsion or braking state.
In algorithmically controlled processes this allows for averaging power input over the entire transport cycle. For manual operator-controlled processes it allows the operator to control the rate of system energy input regardless of instantaneous propulsion or braking demand. The prime mover output can be "averaged" over the entire transport cycle; with the storage system providing peak power. The prime mover is no longer required to respond solely to instantaneous demand defined by acceleration, braking, or grading.