New Qspice update has blocked all my old simulations

I created, thanks to help from this forum, a whole library of old tubes based on tubes models from LTSpice. So far my simulations have worked perfectly. The new update has caused all the diagrams drawn to not work, because with the .tran parameter of 25ms, all calculations in simulations with more than 2 old tubes - when there is more than one, start from microseconds and the diagrams show… nothing for such a short period of time - I don’t want to wait few hours to do that simulation.
The program is no longer useful.
Simulation with one operational amplifier works. The scheme - simulation with two operational amplifiers - does not work - the same situation.

  1. How to turn to previous good Qspice version? Is there any way to remove the timestamp that forces the installation of a new faulty version of Qspice?

  2. Is any solution to repair that problem, because I don’t need very long calculations in microseconds at .tran 25ms




please upload your schematic for problem replication.

I prepare my library - file: TOOLBOX[ZIP].zip archive with all my models.
Please, download that zip archive from https://wetransfer.com/ - only to 05.05. 2025 link - 80MB zip archive. Please, unzip TOOLBOX folder into QSPICE folder in your My Documents Folder

Link to download from wetransfer: Unique Download Link | WeTransfer

Then, create a symbolic link into Qspice to that folder:

Now, we can start :slight_smile:

All my files with “!!!..” are not working - after bad new update they are starting from microsec or nanosec counting and I must wait a long time to see some result - or not working any more.

Before that bad new update everything was working OK.

“OK…” files - are working OK, like before - and it’s strange, that they are working.
.
Now, I don’t know, simulated circuit is not working because it’s with a lot of bugs, or QSPICE randomize that simulation - sometimes works, sometimes is not working.

I need an old previous verion befor 3 months ago - that works perfect for me - every simulation was working perfect. Now - it’s roulette.

Source about Melos SHA-1 amplifire: https://www.youtube.com/watch?v=l0_7LhfnIoA

with some tubes. Headphone amplifire only.

!!!MELOS SHA-1-ON-12AU7-TUBE.qsch (62.7 KB)
!!!MELOS-6N1P-SRPP-PSU-V2.qsch (48.4 KB)
!!!MELOS-6N2P-SRPP-PSU.qsch (46.5 KB)
!!!MELOS-6N3P-6N6P-PSU 250V.qsch (49.4 KB)

Next files not working.
!!!MELOS-6N3P-6N6P-PSU.qsch (49.4 KB)
!!!MELOS-6111-MUFOLLOWER-160V-PSU-V2.qsch (54.9 KB)
!!!MELOS-AN-ECC82-PSU.qsch (48.9 KB)
!!!MELOS-AUDIONOTE-6n1p–PSU.qsch (46.6 KB)
!!!MELOS-AUDIONOTE-ECC82–PSU.qsch (47.1 KB)

OK WORKING-TEST-2xOPA-DAC-OUTPUT.qsch (13.0 KB)
OK-MELOS SHA-1-ON-6N2P-TUBE.qsch (59.2 KB)
OK-WORKING-6N2P-220V-WITHOUT-PSU.qsch (42.3 KB)

Disable “Fast (less accurate) math”. I ran several “not working” schematic and they can all run quickly. However, it becomes very slow if Enable “Fast (less accurate) math”.

MANY MANY Thanks - now it’s working OK!!!

Another question - quest for me. I bought in China PCB for headphone amplifire A1-HD8-Pro and that amplifire is OK for me - now I use it and DC Offset on headphone jack socket is very small,- ca. 5mV only.
I was trying to create QSPICE simulation.
Before I have had created Lehmann Black Cube amp - and simulation with one single OPA is working OK.
Next, I want to try to rewrite from schematic and PCB, circuit to simulate in Qspice - and receive nothing - I have tested many OPA - not working as I expect - with small signal and big DC Offset in the end.

Have you any idea, why I receive a big DC offset in Qspice simulation? Lehmann simulation is OK, but A1 circuit form schematic I had receive from China and real PCB - something bad. But that headphone amplifire is working on my desk.

I was convinced that it was due to this new Qspice upgrade, but when now the tube schematics work - I have no idea what is wrong in the simulation.
!!!headphone-amp-HD8-A1-PRO.qsch (40.7 KB)

!!!headphone-amp-HD8-A1-PRO-v2.qsch (34.9 KB)



OK-WORKING-headphone-amp-LEHMANN-AUDIO.qsch (25.5 KB)

Are you having a mistake in here? AD797 IN+ is connected to -15V!

Thank you very much!!! I agonized for two days looking for a solution, all because of such a stupid mistake! Man, you are great!!!

1 Like

To run with fast math disable, you can add this directive .option fastmath=0 in your schematic and it can override any preference setting.

1 Like