ELEKIT TU-8900 2A3/300B
- Thread starter vkung
- Start date
Cool mods!
...I'm waiting for a Copper top-plate, Copper transformer cover, Copper chassis options
Appears as though you raised the 300B/2A3 4 pin tube sockets, or maybe that is how they always were without mods to the stock kit, IDK, but will say I'm not a fan of amps where tube bases are sunken beneath the top plate.
Do you have or recommend a replacement 12AU7 9 pin PCB tube socket which can easily be installed to sit higher and more flush with the acrylic top plate?
I was looking at these but I don't know TU-8900 or tube socket measurements to say how well they'd work to raise the tubes more flush with the acrylic top plate option: https://www.ebay.com/itm/193493294988?hash=item2d0d195f8c:g:8IMAAOSwY3daZYRF
or a PCB board adapter for CMC 9 pin socket https://www.ebay.com/itm/194037518275?hash=item2d2d898fc3:g:U2cAAOSwW4ddReLw and CNC 9-pin https://www.ebay.com/itm/165383745105?hash=item2681a3ae51:g:OsQAAOSwNuxXYCNy
Or do we have to wait for an updated and revised TU-8900 made to better accommodate the AN caps?
I built a TU-8900 last weekend, and have been giving it extended play since, chipping away at the hours needed for everything to burn / settle in. My build has Lundahl LL2785C, Audio Note Silver Capacitors, Tantalum Resistors, etc. I dropped the silver caps In a few days ago, so I've awhile to go on those too. Overall though, I'm quite happy with the quality of the kit and the sonics I'm hearing (even early on).
I already have a decent 2A3 SET amp and wouldn't mind adding a 300B SET amp to the arsenal. Obviously, I am looking for the highest sound quality, would you steer me more toward the latest version 300B (only) TU-8600S amp with all upgrades or to the TU-8900 2A3/300B amp with all upgrades, and why?
Wanted to take a moment to chime in re: the Audio Note Silver caps. I have about ~80 hours on them, and I'll say this; It was one of the more expensive upgrades, but I'm glad I spent the money. The extra detail extraction (micro / micro) along with further improved separation / staging has reset my expectations on what a 2a3 or 300b amp should deliver. My use is purely with headphones (Empyrean Elites), but the improvements are not subtle.
While they may be out of reach price wise for many - if you can afford them (thanks for the discount @vkung!), they can really elevate this amp (if your headphones / speakers / chain can support the improvements).
While they may be out of reach price wise for many - if you can afford them (thanks for the discount @vkung!), they can really elevate this amp (if your headphones / speakers / chain can support the improvements).
Last edited:
Enjoying my 8900
Lundahl Amorphous (”without form” in reference to random metal grain structure)
V-Caps CuTF
Panasonic Automotive grade hybrid Polymer Cathode Caps
Takman Metal Foil resistors
Ohmite wirewound non inductive Cathode resistors
Cossor E2A3s
12BH7As
Audio Note Kaisei electrolytes on Power supply
TKD pot with diy voltage divider added
Lundahl Amorphous (”without form” in reference to random metal grain structure)
V-Caps CuTF
Panasonic Automotive grade hybrid Polymer Cathode Caps
Takman Metal Foil resistors
Ohmite wirewound non inductive Cathode resistors
Cossor E2A3s
12BH7As
Audio Note Kaisei electrolytes on Power supply
TKD pot with diy voltage divider added
Continued from prior post…
I can’t say this sounds any better or worse than the stock components, as I have no way to make that comparison. However, I find the results sublime!
The burn in time was extensive, possibly due to the use of automotive grade hybrid polymer bypass caps. Likely an unnecessary change. CAUTION! The Bypass Capacitors in this amplifier are NOT electrolyte capacitors! These are extremely low ESR polymer capacitors and are critical to the sound of this amplifier. The use of even expensive high grade audiophile electrolyte capacitors will ruin this amplifier! If in doubt leave them alone!
I have settled on E2A3 tubes and zero negative feedback as my favorite. This is with 10 inch full range lii Audio drivers in diy BLH enclosure. No crossover or baffle step correct circuit—output transformer coil directly to speaker coil. I also tried very sensitive two way horn speakers but found the sound, while excellent, was less controlled—less engaging without negative feedback. for me the two way horn speakers sounded better with negative feedback engaged. Again my favorite was the full range without negative feedback
While not necessary, I back this up with REL sub connected at speaker level. I have tried other subs and other connection methods and this to me is by far the most musical.
Kaisei Caps are tight fit. Had to leave leads long on one resistor and one diode and bend out of way. If you’re doing the build and look at the board it will be evident which ones.
Using a modern 2 volt source I cannot imagine running this amplifier with a preamp even in negative feedback selected. In zero negative feedback mode, you will likely need to reduce the source level through Roon or similar software.
As I am primarily using zero negative feedback and don’t particularly care for roon’s digital volume control I elected to add a voltage divider circuit in front of the volume pot. This is a personal preference and not unique to this amplifier. I have done it with others.
These are Vishay naked Z foil resistors. (Note: the jumper posts are cut on input(top)…look very closely. This actually allows you to use less of the carbon trace on the volume pot (turn control higher) which In theory provides a cleaner signal. Hard to say there is an audible difference, but it certainly does no harm to the signal and greatly improves the functional range of the volume control. I used 67K ohm inline with signal and 24K Ohm to ground (bottom side). If you are using the negative feedback mode, or have a low voltage source, you may not want to do this. Delicate procedure, and difficult to undo if you do not like the results. I did breadboard experiments at the inputs with cheap resistors first. This is delicate work, and certainly not for everyone.
I can’t say this sounds any better or worse than the stock components, as I have no way to make that comparison. However, I find the results sublime!
The burn in time was extensive, possibly due to the use of automotive grade hybrid polymer bypass caps. Likely an unnecessary change. CAUTION! The Bypass Capacitors in this amplifier are NOT electrolyte capacitors! These are extremely low ESR polymer capacitors and are critical to the sound of this amplifier. The use of even expensive high grade audiophile electrolyte capacitors will ruin this amplifier! If in doubt leave them alone!
I have settled on E2A3 tubes and zero negative feedback as my favorite. This is with 10 inch full range lii Audio drivers in diy BLH enclosure. No crossover or baffle step correct circuit—output transformer coil directly to speaker coil. I also tried very sensitive two way horn speakers but found the sound, while excellent, was less controlled—less engaging without negative feedback. for me the two way horn speakers sounded better with negative feedback engaged. Again my favorite was the full range without negative feedback
While not necessary, I back this up with REL sub connected at speaker level. I have tried other subs and other connection methods and this to me is by far the most musical.
Kaisei Caps are tight fit. Had to leave leads long on one resistor and one diode and bend out of way. If you’re doing the build and look at the board it will be evident which ones.
Using a modern 2 volt source I cannot imagine running this amplifier with a preamp even in negative feedback selected. In zero negative feedback mode, you will likely need to reduce the source level through Roon or similar software.
As I am primarily using zero negative feedback and don’t particularly care for roon’s digital volume control I elected to add a voltage divider circuit in front of the volume pot. This is a personal preference and not unique to this amplifier. I have done it with others.
These are Vishay naked Z foil resistors. (Note: the jumper posts are cut on input(top)…look very closely. This actually allows you to use less of the carbon trace on the volume pot (turn control higher) which In theory provides a cleaner signal. Hard to say there is an audible difference, but it certainly does no harm to the signal and greatly improves the functional range of the volume control. I used 67K ohm inline with signal and 24K Ohm to ground (bottom side). If you are using the negative feedback mode, or have a low voltage source, you may not want to do this. Delicate procedure, and difficult to undo if you do not like the results. I did breadboard experiments at the inputs with cheap resistors first. This is delicate work, and certainly not for everyone.
Last edited: