Inthis method, the ballast is the current source for the DC signal. Since the dimmeris the point of reference, this Back End method is sometimes referred to as current sinking. In the 0-10 VDC 4 wire method (also known as 1-10 VDC 4 wire since 1 volt is the minimum level signal), the 2 low voltage wires (violet & gray wires) must be kept separate from the power wiring in the system except at the ballast itself. The control wires are polarity sensitive so accuracy is important in wiring. If the ballast senses a loss of the control signal, the lamps go to full light output. This is described in NEMA ANSI C82.11-2002 High Frequency Fluorescent Lamp Ballasts - Annex A Low Voltage Control Interface for Controllable Ballasts. This protocol definition is also in IEC 60929 Annex E.2. Note that the switch in the dimmer for this application is required to turn off the ballast. [...] This must notbe confused with 0-10V current source applications, found in Section 7.1.
A topology free, two wire, polarity sensitive, parallel bus is connected to all the controllable ballasts and to the controller. Typically, the ballast is capable of withstanding a voltage range of -15V to +15V. 10VDC produces maximum light output, while 1VDC produces minimum light output. Ballasts are protected from reverse polarity. If the control terminals of the ballast experience reverse polarity, the ballast produces minimum light output. If the control unit is not connected, as in an open circuit or control equipment failure, the ballast produces maximum light output. Ballasts are the source of the control current. Controllers for this method are variable resistors or an electronic equivalent capable of handling a range from 0.01ma - 2ma and are hardwired to a ballast or group of ballasts. Typical applications include architectural dimming and daylight harvesting.
This seems straightforward enough. The physical toggle switch at the bottom of the control is a standard SPST switch for switching the ballast power source and can be ignored. The top slide portion of the control is just a rheostat/two terminal potentiometer. The ballast/light fixture uses the 0-10v "DIM+" and "DIM-" control wires to provide a constant current source. When the slide is at the top of the control, resistance is at maximum so the constant current source voltage is driven to maximum (nominally 10v). When the slide is at the bottom of the control, resistance is at minimum so the constant current source voltage is driven to minimum (nominally 0v).
Now for the confusing part. When I measure the resistance of the 0-10v wires with my multimeter, I get 207kΩ resistance at maximum and 29kΩ at minimum. This is likely supposed to be 200kΩ to 20kΩ with some room for trimming (the SF10P-W has a trim knob). The current on a constant current source at 10v and 200kΩ would have to be 10v/200kΩ = 50A. Then at the low end we have 50A * 20kΩ = 1v which matches up with "1VDC produces minimum light output".
How does this dimmer actually work? Is it really just a variable resistor or is there something more complex going on inside? I cited the IES TM-23-11 document as my source for all this as it's the most authoritative and consistent reference I could find but is it wrong or describing something else?
The ballast can contain a very simple current source, not necessarily a constant current source IC. It could be a 600R resistor from a 12V rail, so that when the dimming pair is shorted out you get 2mA, but when the variable resistance is 200kohms it is 60 microamps. The ballast sets its dimming level based on the voltage formed by this divider.
It doesn't make any difference whether it is 100k or 200k at maximum since compared to 600 ohms you will get the full 10V, but it would affect the linearity of control, eg 50% would be where the resistance is 600 ohms.
3a8082e126