EasyEDA Spice simulation needs help!
150 5
eetech00 3 months ago
Hello, I've been using LTspice for a few years now and decided to tryout EasyEDA through one design interation. The spice schematic entry seems easy enough to use but the mapping of symbols to LTspice attribute is confusing. The help document doesn't help one who already knows spice. I'm trying out a simple schematic that contains 4 digital logic parts and a 555 timer. The schematic easily simulates in LTspice but not EasyEDA. When the simulation is run in EasyEDA. a silly message is generated "Your circuit is too complex...blah...blah...blah) If I change .trans settings on the schematic, DEFCON messages are generated in the sim report. If I try to specify settings in the simulation setting dialog, I get another silly message "stop time - start time/max timestep should be less than 10000" that doesn't even make any sense. Someone at EasyEDA should clean this up. It does not help the perception of LTspice which is a very powerful simulator. I can offer some input if needed.
Comments
andyfierman 3 months ago
"The help document doesn't help one who already knows spice." In the interests of trying to improve the documentation, could you explain why and how you think it could be improved? Could you post the link to a public project that demonstrates the issues that you describe? Thanks.
Reply
eetech00 3 months ago
Hello, I've decided that EasyEda will not suit my needs as it still has many simulator issues. But...on the positive side...the interface is nice. I don't have time to submit circuits that demonstrate the following issues but these are easily reproduceable if you have the time. All the models I've attempted to use in EasyEDA simulate quickly and flawlessly on my desktop in LTspice XVII. Some issues: 1\. The ability for spice symbols to hide or show power pins seems to affect simulation\. Output waveform is different\. 2\. There seems to be a problem with using CD4000 models\. I have an extensive CD4000 simulation library and tried to simulate using a CD4013B\. A popup message was displayed indicating a  "pin count mismatch". This was displayed using my own model and spice symbol. I discovered that if I changed the model name from CD4013B to CD4013 (removed the "B") then the model worked without error....hmmmm. There must be a conflict with existing system spice models(?). 3\. The animated LED causes DEFCON messages\. I used my own LED model which simulated correctly\. 4\. If I try to specify settings in the simulation setting dialog\, I get a message "stop time \- start time/max timestep should be less than 10000"\.     This message doesn't make any sense. 5\. It seems that a when a simple 4 component circuit simulation is run\, a message is generated "Your circuit is too complex\.\.\.blah\.\.\.blah\.\.\.blah\)\.     This usually happens when a transient specified duration is greater than 2 seconds. Really? 6\. Sometimes the waveform viewer is not displayed even though the simulation completed\. Good luck with this product. I will try again in a few months...
Reply
andyfierman 3 months ago
@eetech00, Several of your "easily reproduceable" issues are dependent on the specific examples in which you encountered them. Presumably you have created small projects in which these issues because apparent. Unfortunately unless you can make those projects public and share their URLs, it is unlikely that they will be addressed in any detail. **To more generally answer your points:** 1\. The ability for spice symbols to hide or show power pins seems to affect simulation\. Output waveform is different\. That could be because as well as hiding the power pins, the net names and autogenerated supply voltages may no longer correspond to those used when the pins are made visible. Without you specific example, the causes are unclear. <br> 2\. There seems to be a problem with using CD4000 models\. I have an extensive CD4000 simulation library and tried to simulate using a CD4013B\. A popup message was displayed indicating a  "pin count mismatch". This was displayed using my own model and spice symbol. I discovered that if I changed the model name from CD4013B to CD4013 (removed the "B") then the model worked without error....hmmmm. There must be a conflict with existing system spice models(?). Yes, there is an extensive library of 4000 series parts which your model names may conflict with. If your models are based purely on A Devices with no level translation at each end then they maybe do not include supply pins for modelling supply dependent behaviour (such is partly implemented in the late Helmut Sennedwald's libraries and more fully in the EasyEDA 74HC and 4KHEF libraries). Hence there may be a pin count mismatch. There is a known problem in that the 4000 series parts library has been uploaded but the spice symbols do not yet appear in the System Spice Library. <br> 3\. The animated LED causes DEFCON messages\. I used my own LED model which simulated correctly\. The animated LED is intended to be for low speed simulation only as it is can only show low speed flashing. It is not suitable for general simulation use and I have asked for it to be identified as for limited educational use only. In fact most of the spice symbols in the EELib are targeted at educational use, rather than general or advanced simulations. There is a small selection of LEDs in the System Spice library: DLEDR, DLEDO, DLEDG and DLEDY. 4\. If I try to specify settings in the simulation setting dialog\, I get a message "stop time - start time/max timestep should be less than 10000".     This message doesn't make any sense. The actual message is: "(stop time - start time)/max timestep should be less than 10000". The criterion behind the message is a crude method to limit the size of the datafiles produced by a simulation which is a consideration when the simulation is being run in the cloud alongside many other user's simulations. In fact if you simply enter the simulation command directly into the simulation schematic just stating a start and stop time with no max time step then EasyEDA will usually run the sim quite happily. 5\. It seems that a when a simple 4 component circuit simulation is run\, a message is generated "Your circuit is too complex\.\.\.blah\.\.\.blah\.\.\.blah\)\.     This usually happens when a transient specified duration is greater than 2 seconds. Really? It may be that you are trying to simulate a circuit with periodic behaviour that over 2s generates too much data and so is eating up cloud resource. For example a pulse source generating a 1GHz pulse train would likely do that for a 2s simulation run. Without access to your specific simulation however, it is not possible to offer any constructive comment or advice. You can find many examples of simulations including more than 4 elements that run very happily in EasyEDA both in the Simulation Tutorial and in other public projects. <br> 6\. Sometimes the waveform viewer is not displayed even though the simulation completed\. This is probably related to the simulation output file size but which is not being trapped and a suitable error message being produced. You may consider submitting a Bug Report about this wich includes a link to a public simulation that demonstrates this behaviour.
Reply
eetech00 3 months ago
@andyfierman 1\. Noted\. 2\. Negative There is no pin count match and I'm extremely familiar with Helmuts models\. I do use A devices in the models but am careful regarding level translation. I never use models without testing it first. Testing the model is how I discovered the CD4000 conflict. I'm aware of how supply dependent behavior works in the spice subcircuit but could not find any documented information on how the EasyEDA spice symbol interfaces to it other than pin mapping (which I already understand) . 3\. Noted\.  But I think you should go further and mention in the descriptive text of the part that it can cause DEFCON messages so a user will know right away where the cause of the messages may be\. I discovered it \(and wasted time better spent designing\) by removing the LED from the simulation\. 4\. Negative\. It did not run happily\. It let me specify the parameters on the schematic \(10 seconds\) but it produced the "too complex circuit" error\. 5\. Negative\. I was using a pulsed source with a period of 1 sec\. I did enter the duration directly on the schematic \(10 seconds\) but then it produced that silly "too complex circuit" message\. 6\. Noted\. Thanks for you responses. I want to emphesize that the sub circuits I tried to use here have been used for years with no issues. I guess I expected less issues since the simulation engine is LTspice. <br> I'll try again in a few months.
Reply
andyfierman 3 months ago
@eetech00, Your points 2, 4 and 5 require at least sight of the simulations that you are running to make any further meaningful comment. I am quite happy to look at your your simulations and models to see why they are not running satisfactorily. If you do not wish to make them public you can add me to a Team so that I can have a look but as I already said, without sight of them, it is not possible to offer any insights and therefore to use those insights to improve the tool.
Reply
Login or Register to add a comment
goToTop
svg-battery svg-battery-wifi svg-books svg-more svg-paste svg-pencil svg-plant svg-ruler svg-share svg-user svg-logo-cn svg-double-arrow