Diode circuit not working

[rachelgl]

Hello all!

I'm having a problem getting my simulated circuit to translate into real life.

Please see attached circuit or image of circuit. I'm checking the voltage at the point "test". When I run a transient simulation, I get a square wave voltage at this point which alernates between 24V and (roughly) 10V. When I build this circuit on PCB, the voltage at this point varies between 24V and 0V. Also, in the simulation, the voltage at "test" relies heavily on the load capacitor (10pF), but for my real life circuit, this capacitor does not seem to have any bearing on the voltage.

Can someone help me to understand this?

Thanks!

[johnsold · Accepted solution]

I cannot run your simulation, but I can offer some suggestions based on the images.

In simulation or analysis of the circuit I would expect that on the first pulse, the capacitor will charge to the source voltage (or very close to it, depending on the exact diode model used). After that the voltage at Test is undefined for ideal diodes (perfect conduction when forward biased and no conduction when reverse biased) because Test is isloated. However Test will always be within the range of 0..24 V because outside that range one of the diodes would become forward biased in at least one state of the source.

In a real circuit the input impedance of your measuring instrument becomes part of the circuit. If you are using a digital multimeter or an oscillscope with a 10X probe, you probably have an impedance to ground of about 1E7 ohms resistance and a few picofarads of capacitance.  The resistance could easily be much lower and the capacitance much higher with other equipment. With this impedance you have currents flowing to ground through D1 when the source is at 24 V and leakeage currents through D2 when the source is at 0 V. Since the capacitance of the measurement system is comparable to C1, the effect of C1 will be hard to observe.

Lynn

[rachelgl]

Thank you for your advice. I will re-try the measurement of my device and see if I can get different results with different equipment.

Thanks again!

[digitalguy]

When I put a 10 MegOhm resistor from the test point to ground, to simulate the impedance of a typical o'scope, the signal does vary from 0v to 24 volts as your breadboard circuit does.