• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

uTracer6, ExtractModel and getting obscure error "*** Internal error in DetSweep"

Having used a uTracer3+ and ExtractModel for years to create spice models now I am trying to create spice models with a uTracer6 as the 6P36S can support very high plate currents that I want to model in spice.

I am using the ExtractModel software from DOS4EVER that I have successfully used for years with my uTracer3+.
Now with the 6P36S tube data collected with a uTracer6 I get the error in the title, "*** Internal error in DetSweep".
I have so far failed to find the cause.

Does anyone have a list of error codes for ExtractModel that may provide a hint as to what it is not liking in my data?

Has anyone see this error before?

I attached the data files I am using.
 

Attachments

Unfortunately nothing on this issue on the google groups.
I have hoped to avoid taking up Ronald's valuable time before I tried other avenues but now I have put a question to Ronald.
I will post any response I get in case it may help others.
 
I received a reply from Ronald.
Sadly due to family difficulty he is unable to provide support right now.
Ronald has made a excellent contribution to the community with the wealth of information on his WEB site and the creative and interesting designs he has published.
I send him and his family only the best of wishes for a better future.
 
Hi, anyone have contacts of utMax GUI and firmware for uTracer? If yes please PM.

I recently contacted Nick by email at black magic amps.
There is no firmware update for the uTracer6, yet. Not sure if or when it may ever come. Hope it is sooner but seems a long shot.

You could email Nick at black magic and ask if you can still purchase PIC chips for uTracer3+ with improved firmware.
Black magic web site states the PIC as "sold out" so it would be a favor to sell to you one it seems.

There is a Linux open source version of uTmax on github.
https://github.com/skullandbones/uTmax

Unfortunately I am not enough of a programmer to fully understand how or if a PC Win version can be created from the provided source.
Then again Linux can be real nice compared to the ongoing Winxx parade.
Creating a executable for a PC is however a bit above my current pay grade. I would need clear instructions on how it is done and the tool chain setup required.
If there is someone out there that can post a "how to" that would be really great.

I can tell you the PC version of uTmax does fix a few of the accuracy problems I know are present with the uTracer6 and Ronald's GUI.
Since Ronald unfortunately does not seem in a position for update his GUI, uTmax may be the best way to go at present.

However I am currently concerned and thinking about if investing more of my time in my uTarcer6 is a good thing as it is unclear where if anywhere it may be going as a product.
I purchased my uTaracer6 as I believed it would be basically a "turnkey" low time investment way to acquire a more capable tube curve tracer over my uTracer3+ that over the years I found to be excellent. I just wanted a improved more capable utracer product not a long term project.
The accuracy issues I have found right from the get go with my utarcer6 however has turned it into a bit of a time sponge.
The uTarcer6 is still not a usable tool for the purposes I purchased it for. That was testing high current sweep tubes (up to 1 amp) at lower voltages (less than 100V) and to get accurate results. Contacting Ronald about the short comings I found in accuracy of the Utracer6 at low voltages and high current did not result in any solutions. Really too bad as I think the uTarcer6 is real close but not quite there yet to be an useful accurate tool at higher currents and lower voltages.
It works fine at higher voltages.

If the PIC and GUI code were open source I would feel very differently about my uTracer6 as then I would have full confidence issues and improvements I have identified in my utracer6 i could address. Micro controller code hacking I can handle and fixes for the issues I have encountered would/should be minor changes.

There are now open source arduino based curve tracers out there that are seeming like a possible better investment in my time than continuing to hit my head on the closed source wall where needed fixes and improvements may never happen.

Let me be clear.
I like and respect Ronald and think (know) he has made a great contribution to the community and none of the above should be seen as a criticism of him in any way.
 
Last edited:
A update on uTracerJS.
With the uTracerJS GUI and my utrace6 I can acquire accurate data at high anode and screen currents and low voltages (less that 100V) without data corruption.
ExtractModel can then process the resulting data files and create a spice model without error.

There is a know bug I uncovered in Ronald's GUI with the utracer6 that can result in corrupt data under test conditions of I(Va=Vs, Vg) with Vh constant sweep.
Ronald has not yet issued a update for this issue .
I have confirmed in testing the utracerJS does not corrupt the data under the same condition that Ronald's GUI seems to.

So I will speculate for now that perhaps the ExtractModel error I encountered is the result of using a Rondold's GUI and utracer6 to collect data and the failure in ExtractModel is a result of data corruption.
Going forward I will use utracerJS with my utracer6.

See https://boffin.nl/wp/ for utracerJS
 
Last edited: