Reducing the uncertainty on solution exchanges by monitoring solenoid valves actuation in real time.

JERÓNIMO AUZMENDI; LUCIANO MOFFATT

Huerta Grande, Córdoba, Argentina

Congreso; 1º Reunion Conjunta de Neurociencias (IRCN); 2009

Sociedad Argentina de Investigación en Neurociencias (SAN)

Usually, the effects of chemical compounds on biological preparations are tested using solution perfusion systems. Solenoid valves are a core component of most solution perfusion systems used in neuroscience. They control the timing and sequence of chemical stimulation. These valves have a ferromagnetic plunger which moves after the magnetization of the solenoid and which returns to its resting position by the aid of a spring. There are delays between the time of voltage application or removal and the actual aperture or closure of the valve, that are difficult to predict and have to be measured experimentally. We define as plunger signal an electric signal that is generated by the motion of the ferromagnetic plunger and that is measured over a small resistance that is set in series with the solenoid coil. We tested a 3-way pinch valve and a 2-way solenoid. In order to test on real time whether each valve was open or not we set the valve to connect a pressurized nitrogen tank to a differential pressure sensor. When the valve was open, the pressure was 0.15 bar. When the valve was closed, the pressure was equilibrated with the atmospheric pressure with a time constant of about 1 ms. We used the multifunction data acquisition card NI PCI 6229 for synchronously driving the valves and measuring both the output of the pressure sensor and the current that circulates through the valve coil at 250Khz 16 bits digitally filtered at 10KHz. The circuit driving the solenoid valves consisted of a measuring resistance in series with the solenoid valve in parallel with the series of a protective diode and a 47V Zener diode. Using signal´s plunger we detected the opening and closing of the valves with a systematic error below 2 ms. After this systematic error is subtracted, the trial to trial error was below 0.2 ms. Because there is a strong interest on the ability to determine the delay between the application of the voltage signal to the valve and the time when the valve actually opens, we propose a simple way for monitoring in real time when does the solenoid valve open and close.