On Mon, Nov 18, 2024 at 11:33 PM, <dhara.weerasinghe@...> wrote: ... I get "time step too small" error. Could you please direct me to a thread that addresses this problem. For Timestep Too Small

Adding GSHUNT seems to have worked for me. The shift in levels in longer simulations makes sense.

Don't overlook that in the drop-down for "Top Directory" you can use "./../.." and similar to create hierarchical tree with your symbols at the lowest level and various projects in branch folders. Thi

john23, I don't know if this helps, but I will offer these suggestion: In your schematic, I recommend connecting V1 (the 17 V power source) between the VDDx pins and the VSSx pins. In other words, not

Hello All, My circuit has a top page schematic and four sub-schematics. By replacing each sub-schematic with a dummy behavioral circuit, I was able to simulate fast. After working through the 16 combi

Hi Andy, thank you for your replies and suggestions. I must be in a different timezone, I wrote just at business-hours end and now it is business hours start on the next day. I have also uploaded the

Sorry, I missed the K statement and that the circuit is supposed to be a SEPIC. First comment still stands. On 11/08/2023 00:55, Tony Casey wrote:

Sorry, I just quickly read off the datasheet the first Eon entry I saw in Table 4, where it says "Freewheeling diode = MSC040SMA120S (Vgs = -5V). However, the diode mentioned in the section below (MSC

Hello Andy: Thanks again. Udo made SEPIC PFC & ra it successfully. So my closed loop circuit has some errors. Will explore it today with Udo's suggestions incorporated. Will post the file if I get the

Ok. See https://groups.io/g/LTspice/files/Temp/OptiMOS5_80V_Mosfet_V2.zip

For experimental simulations, some alternatives to LTC6752 for comparator U1 include: (1) UniversalOpamp2, with altered parameters to make it faster (much larger GBP and Slew). Op-amps may not make th

Bingo! I so rarely change defaults that I'd forgotten the Alternate solver. And I _always_ reset them when I'm done, right? Donald.

re: "https://ltwiki.org/index.php?title=Convergence_problems%3F" I think you must use the "http" version to access the LTwiki: http://ltwiki.org/index.php?title=Convergence_problems%3F YMMV and may de

"Just shows that sometimes using "real" models and not "ideal" models makes a difference." Yes - the "ideal" diode models are known to be a problem occasionally, for this exact reason. It makes them n

Vlad wrote, "Have you tried with the expression from the comment instead of the active one?" I didn't. I thought it was there just as a comment, something that had been used previously but not meant f

such might even pose a risk for the real circuit some oscillators set demands to their supply impedance ... which might be strange or bizarre if you first meet such dependency ... (also may affect the

I tried the MCP6541 model Andy referred to a while ago and still saw "timestep too small" issues (I went through every step in the FAQ trying to fix it to no avail). Eventually I gave up and used an L

Jorge wrote about a "timestep too small" error. Apparently, Jorge either found and fixed the problem himself, or ran into further problems, because the message was deleted and all the files were delet

The best resource I've seen about fixing "timestep too small" errors is this one: Files > z_yahoo > FAQ > faq_17-2.txt https://groups.io/g/LTspice/files/z_yahoo/FAQ/faq_17-2.txt It has a large section

Try adding this function to your schematic .options cshunt=1e-15 This will add a very small amount of shunt capacitance to all of the nodes to ground. 0.001pf in this case Tony Casey wrote on 6/11/202