From Sam F: see docs/bug.txt 8-6-2007 "It would be great if you could add a fuse in the circuit of the simulator. Also, if we could choose between the real or the conventional direction of the current." Notes from Sam M, 7-29-2007: If you save a circuit with a closed switch, when you load it back up, the switch is open but electrons are flowing through it. If you close and reopen the switch, everything behaves normally. I'm guessing this is because the default graphics for a switch is an open switch. Notes from Sam Reid 7-16-2007: o There is a delay on bringing out the 1st component from the toolbox. o When measuring the voltage across a battery with the voltmeter, and proposing a match from a wire, voltmeter temporarily reads out infinite voltage. o Switch handle should give a different mouse cursor to signify that the switch can be thrown. o Example at http://www.physics.uoguelph.ca/tutorials/ohm/Q.ohm.example.parallel.html is handled correctly. After reviewing the version of CCK on the website, it looks like it does a much better job of handling the values shown when editing capacitance, inductance, and voltage, and also of resizing, than the latest version. In the web version, the circuit stays centered in the same place when you resize it, instead of jumping all over the screen. Staying in the same place seems preferable. Also, I don't have the problem of the voltage meter not working after resizing in the web version. So in my opinion, these issues should be fixed before posting the new version. Thanks for this bug report. Save/Load appears to work only if the user specifies the file to end with ".cck" Can you use this as a workaround for now, and we'll try to address this bug when we start development again on CCK in a couple of months. Let me know if that workaround works on your machine. *I have developed several circuits using the Circuit Construction Kit for DC and AC but am unable to save them for later use. (The save button does not seem to work) It would be of great value to save the circuits. Do you have any ideas how I might overcome this problem.* >Hi PhET ... we love your simulations here at St. Ignatius in San >Francisco and use them a lot! >My students are doing a lab using your AC & DC circuit simulator where >they set up an LC oscillator circuit with a 0.02 F capacitor and a 10 >Henry inductor. The students are calculating a period of >2.8 seconds but the simulator puts the period at 28 seconds when >measured with a current graph or virtual stopwatch. >Have you seen this problem before? Am I missing something obvious? >Thanks! Dear PhET team- I was using the Circuit Construction kit with my students today, and had them testing the items in the grab bag. Several of them pointed out that the pencil did not behave as they expected. We had done a lab early where they used pencils as electrodes for electrolysis, using only 3 volts. Thus, they were surprised that the pencil does not appreciably conduct at low voltage. Why is that? ################################################ #To Do ####################### (not fully pruned) · I think that I have found a problem with the capacitors in your AC/DC simulator. The values that I obtain are high by a factor of more than 17. My procedure is to charge a capacitor (0.02 F) with 27.1 V. The switch on the battery side of the circuit is opened the capacitor discharges through a 100 ohm resistor. Measuring the time for the voltage to drop to 10 V should yield the time constant RC (100 ohm x 0.02 F = 2 s) for the capacitor/resistor pair. I get a time constant of >35 s. Chris's bugs. -Bug 12: Not worried about this for short term. -Bug 15: FIXED on my computer (?) -Bug 16: Also a problem in current version online -Bug 25: Needs to be fixed at some point ... old version doesn't have this problem but is worse than current problem -Bug 26: Good compromise. -Bug 33: Also a problem on current version online Kathy's New bugs/fixes: - the non-contact ammeter is ~50% larger than the old CCK version. Please match the old size. - I like the look (font, slider) of the voltage adjustment window and right-click drop down menu on the old CCK version better than the new version ... was that change associated with going to picolo? - Switch needs to be *in front* of wires ... otherwise the switch gets obsured by wires. (it is correct on old version). In old version, wires were also behind bulbs (always) where as now they can obscure bulb. - When has complex circuit... sim doesn't refresh correctly when "unhidden" ... that is when minimizing a window that is in front of sim to reveal sim. (Only part with moving electrons refreshes under these conditions). - When sim has a lot of circuits going ... the right click window does not get drawn correctly ... nothing appears in the window. - The electron speed is different from old to new sim, but I think it is OK. (New sim has slower electrons). - Is there a way to have the computer detect when the sim is completely obscured by another window, and then make it stop running at that point ... right now to get it to stop you have to minimize. I’m having some problem saving sim files, though. No dialogue box appears to save, or re-save files with. Could this be a bug in the upgrade? Thanks a lot for such a work, my pupils would be very happy with it. But as I am concerned, I can't make the voltmeter works and simply have the like-life view but I don't have any access to the schematic diagram. (1) Have you considered adding a diode and a LED to the set of components. These devices are on all Physics courses in Ireland and I would assume in most other countries as well. (2) Labelling the terminals of the battery as + and ? may help students (and some teachers) resolve difficulties about current flowing one direction and electrons flowing the other. It will also help students to identify the terminals on a real battery. (3) The voltmeter / multimeter is excellent but only one can be used on the screen. If 2 or 3 voltmeters could be on screen at same time this would help me explain potential divider circuits better. (4) You might consider adding another metal and a piece of plastic to your grab bag to reinforce the idea that metals conduct and non metals do not. I hope that these ideas are of some use to you. Sam, Alex was playing with CCK yesterday and figured out how to make current flow in an incomplete circuit. It's reproducable but requires a specific combination of times. The attached screen shot shows a circuit that had electrons flowing up towards the top and you can see the bulb was lit. When reproducing this I found that the current would not flow until the switch at the top was partially closed. If closed all of the way, current stops flowing. Once the switch had been partially closed, to begin the current flow, then it could be opened all the way as shown here and the current would continue to flow. If I remove the to wire at the end, current did not flow. If I removed the resistor at the top, current did not flow. Not sure how important this is to fix. It's very bad but requires an exact setup to create the problem. Wendy Sometimes after disconnecting a wire then "split junctions" there is a floating junction. From: Marina Milner-Bolotin To: Sarah.Mckagan@colorado.edu References: <45ABD31B.1070808@colorado.edu> Hi Sam, I actually had the same problems. I didn't notice a new simulation till Carl showed it to me. Probably when we are used to something, we do not notice the changes. I also noticed a few things about the AC simulation, which I found a little uncomfortable or which I wish could be improved. a) The graphs in my view are not very convenient to show the phase shift between current and voltage. They are not connected and since the time is "running" I cannot see the entire graph and I cannot figure out how to Auto scale it. When I choose too large of a value for the voltage of the battery, I couldn't adjust the graph scale. I couldn't figure out how to scale the graph - change axis value (for the vertical axis). b) I could not also figure out how to make the graphs bigger so if I use them during the lecture everybody can see what I am trying to show. Thanks, M. High priority issues that weren't yet addressed: 1. Capacitor physics issues. 2. Current chart shows inconsistent sign of current. 3. Performance issues on some machines. Top taskbar is covered up on windows. >>Should I do anything about this? Code is to maximize to visible screen area (discluding taskbars) for <=768 screen height, >>otherwise center at the top of the screen, leaving 100 pixels at the bottom. Fixed: When dragging a junction in a connected circuit, there becomes a gap in the electron flow 2 components away. >>My fix has the effect of holding the electrons still while dragging a junction. >>I could make them continue to flow with constant density with a bit more work, but not sure it's worth it. Solution for "Fixed: When adjusting a voltage or resistance continuously with a slider, the circuit doesn't update until stopped dragging." includes a non-elegant hack; not sure of an elegant way to solve this problem. "Advanced" controls should be show/hide, not enable/disable When Steve used CCK in class yesterday, some students interpreted the "Disable" button in the advanced section as "disabling" or "turning off" wire resistivity. This is a reasonable, but wrong interpretation, and it might be better to change the button from "enable/disable" to "show/hide" to avoid this confusion in the future. Voltage chart reads wrong value after switching sizes Chart tips don't move, should move on resize Inductor & capacitor incorrectly readout ohms, not Henries & Farads Change in AC Frequency shows readout for voltage. Noncontact ammeter doesn't update its reading when wire resistivity changes. He had a circuit partially built, then tried to undo something by reflexively hitting Ctrl-Z. His circuit dissappear, and was replaced by a battery and a resistor with electrons flowing in both directions through them. I tested this on my PC, and the same thing happened. This test behaviour should be taken out. He would like to be able to reduce the number of significant figures in the ammeter so that the resistance in the wires doesn't keep the numbers from looking perfect when he constructs simple circuits. He said that when he first builds a simple parallel circuit, he would like the currents to look exactly the same in both branches. With three decimal places, they don't, and he has to explain it at a time when he'd rather not. Get the time units to work out correctly. o Is the ac current (in a circuit with capacitor) symmetric? o If not, is the asymmetry correct? -When you have an AC voltage source and a light bulb in series and turn the frequency to 10 Hz, it looks like the current drifts slightly (counter clockwise) as it oscillates back and forth. I put the mouse cursor next to one of the blue dots and watched how one particular blue dot moves over time. Every 10 secs or so, the current stutters ever so slightly and shifts the current over. This seems like a very small/minor problem, but I'm not sure if it's effects can be easily seen with a more complicated circuit. -When I have 2 caps in series with a bulb and battery and turned the voltage up rather high, only one of the caps catches on fire (the one near the battery). See cck_fire.gif. >>Couldn't reproduce this yet. -There's a slight issues with the capacitors. I charged up two identical capacitor at different voltages (100 V and 10 V) so that there are different amounts of charges on them. Then, I removed the charged caps from their circuits (see cck_caps_separate.gif). I connected the positive side of one cap to the negative side of the other cap. I expected to see the charges on these two connected plates redistribute themselves, but nothing happened. (see cck_caps_together.gif). When you connect the outside plates of this 2 cap combo with some wires, the charges on the plates then properly distribute themselves. >>Yes, this is a real problem I'm not sure is addressed by my circuit analysis. I'm not sure whether my circuit analysis should be giving the right behavior here, or if we'll have to detect it and handle it separately. -On that same note, when I had 2 independent charged capacitors (not attached to anything else) it was sometimes hard to grab them with the cursor and move them. When you click on them, they would not be highlighted in yellow like usual. - Unfortunately performance is an issue ... it's pretty easy to get to the point where the mouse isn't working well and it is frustrating to build anything... is there any hope of improving this (ever)??? -I noticed the same performance issues as Kathy. As soon as there were many components in the play area, things got bogged down. Is this fixed in the piccolo version? - is there any way to make the current graph smoother? Right now it shows features that aren't really there? What should happen to the charge stored on a capacitor while the user is editing the capacitance? - Changing capacitance still doesn't show the right thing ... it doesn't show the electrons flowing again to fill up the plate with more or less charge. >>Yes, I thought it would be awkward to keep the charge constant across the capacitor while editing the capicatance (which would be implemented by manipulating the potential drop to account for the difference). The potential drop is constantly changing, based on other elements in the circuit, and I'm not sure of a good way to account for both of these issues. - I ran across a weird bug, (see attached). I had a nice looking circuit and when I was trying to add a bulb to it, it was sort of bogging down as I was dragging the bulb, and then all of a sudden it collapsed to one line, as shown and totally changed the components in the circuit (?). Do you have a short cut of keys that I may have pushed? >>I hope this problem will disappear when CCK is moved to piccolo. Has this problem disappeared in the piccolo rewrite? > We found a bug in AC cck. See the attached screenshot. To reproduce > it, hook up the circuit so it's all attached in a circle, then split the > junction that's shown split in the screenshot. See screenshot in email. There is something I noticed a while ago and forgot to mention. On CCK there is a clear button and it’s in a tool cluster labeled ‘File’; however, it’s job is that of reset and brings the sim back to it’s initial setup. Can you call it reset and move it out of that tool cluster? It came up today because it caused confusion for an individual using it in the booth. >>Asking for clarification. Circuit does fit better. I’m not crazy about requiring labels on all these clusters however on Advanced, it is necessary unless you replace Enable with Advanced. Deleting these cluster titles may be especially helpful in AC CCK if you’re running out of room. What do you think? Retitle the top tool cluster from "File" to "Circuit" in the control panel. Froze at Merlot talk. Bug report from McKagan: open circuit still conducts in particular situations • There is still a visible voltage drop across a plain wire, this should be removed. ################################################ #New Features ####################### See faraday's AC source graphics? Quantitative graph values. Quantitative CCK can wait until after AAPT - graphs ... would be nice to add dashed grid in background so easier to read where you are, also would be nice to have a cursor which could move around and scroll along graph, this would require a play, pause, step button to be added in bottom of sim. • Graph controls (quantitative) o Cursor o Readout While working on the interview paper Carl thought of a possible addition to CCK. What do you think about adding a person who can be used to test out the physiological responses to electricity. I’m not sure of the easiest way but he would have various reactions according the the stimulous. If it were as simple as touching a battery to his tongue (maybe to hard to integrate into CCK) he’d just jump a little and say something appropriate, as the shock gets stronger he may say ouch and finally it might just turn into a pile of black ashes or something. Problems to think about are making it too much fun and CCK is already very full. Possibly put him in the advanced panel. We came up with this because one of the benefits of simulations is being able to take risks without negative consequences and when students study electricity this is one of the most engaging topics and in lab students are quite often afraid of electrocuting themselves. What do you think? ################################################ #Lower Priority ####################### Help button should be moved south. o Improve the approximation for determining capacitor current. -Could use solution from companion model, but more difficult to code. There is a noticeable delay when dragging a component from the toolbox for the first time; this has confused some users. Bug: capacitor clips are clipping all electrons, not just electrons in that branch of the circuit Explain and remove fudge factor for capacitor in MNASolution? -If you build a simple circuit (battery, resistor, capacitor, and wires in series) and wiggle an element or node back and forth around with the mouse, the blue balls stop flowing from a node 2 or 3 nodes clock-wise from the element or node you are moving around. The blue balls stop flowing through this particular node, but the rest keep on flowing, so you get a wire or element with an absence of blue balls because they all flow out. This may be due to the fact that the wire has to redraw the number of blue balls because wiggling an element or node changes the number of blue balls (according to its length). o Set node voltages from the MNA, then use that to compute voltmeter readouts (instead of complex and error-prone graph theory computation). Quantitative numerical tests for RC, RLC Circuits, etc. -When there were two circuits running in the play area, there was definitely a decrease in performance and things ran a bit slower. If you grab a wire, say, from the tool bar and drag it around in the play area, there is a noticeable lag between the cursor and the wire moving with it. -When you have 2 caps in the play area and you turn one of them 180 degrees around, it looks a little strange next to the other one that has not been turned around--sort of an MC Escher effect. See cck_escher.gif.