That looks like it worked. I moved the archive file to the same folder where the SIT3X-Build-Guide folder is located. It successfully unpacked the new files and replaced the Outline text file after a query.
I have created a github repository for the SIT3X Build Guide.
GitHub - lhquam/SIT3X-Power-Amplifier
Download from there. Let me know about success or problems.
GitHub - lhquam/SIT3X-Power-Amplifier
Download from there. Let me know about success or problems.
For first-time GitHub users, once you arrive at the page click on the "Code" green button and select the option "Download ZIP".
It says "Code' because the main purpose of the site is to store software source code and associated files.
🙂
It says "Code' because the main purpose of the site is to store software source code and associated files.
🙂
I have a maximum of 2 sets of boards remaining.
I can send you a set. Send me the your shipping address and make a Paypal payment of $50 to email address lhquam@protonmail.com.
I am still waiting for ml48 to indicate whether he want boards.
Lynn
Thank you for this great approach. I can download the zip, extract the file structure with directories etc. Many of the jpg are there. However the BOM and xls files are links through ../BOM and ../sim directories that are missing and thus so are the files themselves.
Any others have the same experience? Or did I miss something?
Pierre
Thank you, I was going to check that but didn't get to it yet. It will be fixed soon.
I believe I have replaced all of the symlinked files in the git repository. Let me know of any remaining problems.
On the front end, is there a suitable alternative for M1/M2 (2Sk2013 and 2SJ313)? I am having trouble finding these - or are there some reliable vendors anyone would recommend?
I see in other similar applications people are substituting Fairchild FQP3N30 and FQP3P20. Would these be suitable in this design?
I see in other similar applications people are substituting Fairchild FQP3N30 and FQP3P20. Would these be suitable in this design?
In my SIT3X prototype I tried to use FQP3N30 and FQP3P20 FETs, but was not satisfied with the performance, mostly because of the higher Vgs of those parts. The Vgs problem gets worse as the rail voltages are reduced to lower the heatsink power dissipation. A front-end more like the BA3-FE might be more suitable, but then we will have a positive phase second harmonic unless the front-end deliberately has very high negative H2 phase.
If I was to order a matched pair of Tokin 2SK182ES from watanabe or pras what parameters would I need to request?
Boards arrived safely today. Thanks so much Lynn for sharing this design and going to the trouble of offering the boards. Now to find the pair of 2SK182ES I’ve got stashed away somewhere!
I've also received the PCBs. Thank you Lynn for generously sharing with us
your design and making the boards available.
Cheers,
Dennis
your design and making the boards available.
Cheers,
Dennis
That is a good question that I cannot really answer. Vgs isn't particularly important because it is adjusted by a pot. What is important is the behavior (transconductance and harmonic spectrum) at a particular operating point defined by 3 parameters: Vds, Id, and Rload. Both Id and Vds are pot adjustable, but Id is limited by allowable heatsink power dissipation. Watanabe will not really be able to select for the imporant characteristics.
The good news: I had six 2SK182ES SITs available, two of which were selected for my SIT-2 style mu-follower amplifier. I selected from the remaining four based on measurements described in https://www.diyaudio.com/forums/pas...plifier-build.html?highlight=sit+measurements. There SIT are nowhere near perfectly matched, but the SIT3X OS seems to be quite forgiving of device mismatches, particularly with the adjustments that are available.
Within a given part, several methods give decent matches. Vp is a good one,
or Ids with a given load like 8 ohms is also workable.
or Ids with a given load like 8 ohms is also workable.
I am not sure I understand what you mean by "Ids with a given load". Wouldn't Vp and Ids depend on Vds also because on the triode behavior of the device?
Have you found that by selecting using those parameters that other device characteristics tend to "correlate".