Yes, it would have a big input on such a low voltage signal. OTOH if you use a high enough source resistance, the current will be limited to a safe value for the internal protection diodes anyway. That's because I edited my reply just after posting it, because I misinterpreted what your simulation was showing - sorry!Īnyhow, you brought up some very good points about startup conditions and whether the BAV99's can actually do anything to protect the Atmega input pins.īAV99's will take some of the current away from the Arduino protection diodes, but what proportion they take is not very predictable. It's the weirdest thing, my iPhone is showing me a longer reply from you earlier, but the web site does not. Makes me reconsider some of the other outputs as well! Perhaps time to return the product in favor of the other circuit analysis tool on offer in the App store. Interestingly, iCircuit reports a signal across a NO standard relay being possible even if the coil is not energized. Based on the new reply from you (that popped up while I was writing this) I will retain the 100uF capacitor. I tried out the circuit with components that have a 10x higher resistance and naturally that works also. The question is whether a triac or SSR would have an impact on the signal. detect a high voltage condition and turn the thing off. Alternatively (and perhaps smarter) tie the gate to a digital pin so that the analog input into the atmega can be interrupted at will - i.e. Tie the gate to the 2.5VDC power supply via a 10k and the thing should only conduct whenever there is AREF power. One solution I thought of (but cannot implement yet thanks to the iCircuit program being on strike again) is using a small triac or SSR in series with the coupling capacitor separating the transformer output from the Arduino. Anyhow, you brought up some very good points about startup conditions and whether the BAV99's can actually do anything to protect the Atmega input pins. There will also be a small phase shift due to the magnetising current of the transformer and the resistance of its primary. This is not a problem if you are limiting the input current to a sufficiently low value. The voltage at the Arduino input pin will still go negative at startup with that design, until the capacitor has charged sufficiently. You could also remove the lower resistor (currently 1K) in the voltage divider if you increase value of the upper one (currently 2K2) by 50%. The 22 ohm resistor does nothing useful and might as well be removed. I would increased the values of all the resistors by a factor of 10 or 20, to reduce the effect of the capacitor on the phase shift, and also to reduce the current through the protection diodes when transients occur on the input. A lower value capacitor will cause more phase shift. So you need to minimise the phase shift in the voltage you are sampling. If you are trying to measure power by monitoring voltage and current, then you need to take account of the phase between the voltage and current. I also got similar results when I omitted the transformer and used a AC signal source directly on the resistors that would have been attached to the secondary side of the transformer.ĭo you think I should stick to the 100uF ceramic capacitor or breadboard the circuit above on a existing Mini that has already lost one ADC channel due to earlier mishaps with voltage dividers?Īlthough reducing the 100uF capacitor to 10uF only reduces the voltage slightly, it causes extra phase shift. The DC was still kept out - note the 100VDC offset I put into the power supply. using a 100uF coupling cap in terms of transferred signal. That aside, I thought it was very interesting that there was only a very small marginal difference between using a 10uF coupling cap vs. So the current flowing through the primary is somewhat fantastical. This one includes the transformer though I have to admit that I did not include the internal resistances of the transformer because they were not listed by the manufacturer on the datasheet. Kinda defeats the purpose of the save function, no?Īnyhow, I thought I'd include another screenshot. Worse, the current version on my CPU will allow me to save designs but not to re-open them later. OK if the signal amplitude is similar, not so great if you're trying to monitor two very different signals at the same time (i.e. For example, only one scope window can be opened. Don't get me wrong, it has a nice GUI, a number of elements that can be added easily (everything from simple circuit components up to ADCs) but there are several improvement opportunities. I downloaded iCircuit from the App store and I am somewhat underwhelmed by it. I tried out a couple of tools yesterday, including MacSpice.
0 Comments
Leave a Reply. |