• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

After a 14 year run, the TSE must DIE!

A slightly bigger board would be good - extra space for things like bigger heatsinks for the MOSFETs (those things run very hot). But then a two board solution would be good as well, taking the power supply on to separate board would give more flexibility and make it easier to experiment with alternative power supplies.
 
Today's plan is to grow the board up to a maximum size of 5.75 inches by 9 inches.

bigger heatsinks for the MOSFETs (those things run very hot)

One of the "minor changes" is to move a lot of that heat into a big resistor or two, so the mosfets wont burn your fingers.

Another common request is to add the screw terminals like seen on the TSE and SPP. I will also add terminals for an external power supply if there is room for it, otherwise the existing pads will have to do.

C1 needs to be a lot bigger to accommodate 2A3 tubes.

A clamp diode needs to be added across the B+ supply. The supply goes negative at power on until the rectifier tube warms up......can't be good for the electrolytic caps.

Virtually nobody except me used the PC board mounted volume pot, so it's gone.

Library part for the 5842 tube is being updated to the same footprint used in all other Tubelab products......the TSE board predates the existence of Tubelab Inc. It was actually named after my prototyping system I called "The Tubelab."

I will be leaving on a 12 day road trip on Monday (weather permitting). During the down time I plan to do the layout work, and order parts, so I can make a prototype board when I return.
 
What about some jumpers in the cathode leads of the power tubes? Installed, they connect the cathodes to the onboard supply. Not installed, they provide pads to attach an external heater supply for 6.3 or 7.5 volt, or other, tubes.

Would need one jumper for each tube so they are isolated from each other.

Win W5JAG
 
I had a box full of PCB mounted dual pots that I got surplus back in the early 90's. They were plastic, but worked good, and none ever failed. I had found the same parts in one of the popular catalogs of the day and falsely assumed that I could get them when needed. When my supply ran out in the mid 2000's, none were to be found.

The picture in post #34 shows my work-around. It is a common Alpha pot from Mouser with wire looped through the holes in the solder lugs on the pot and then soldered into the board. I used this method to build several amps, but I haven't made complete amps in about 10 years. I assume that the rather standard pots that I used back then are still around, but I haven't actually looked for them.

I am away from my lab at the moment for an unknown time frame. I probably still have some of them in the original bag. I will check when I get back. It could be 2 days, or 2 weeks, probably somewhere in between.
 
Seems like just a few years ago I bought them for under $10 each.......I guess I should sell mine then.

I think I paid $15 each for mine nearly 10 years ago. I could sell them, but I only have two and I would find a need for them right after I sold them......I did sell off my collection of Telefunken tubes last year, then used the funds to buy some "worthless TV tubes." OK, a lot of TV tubes, all for $1 each or LESS.
 
First Prototype assembly started. I spent a little bit of today soldering parts, and the rest of the day ordering $350 worth of components from Digikey. The Mouser order may get done tomorrow.
 

Attachments

  • P2360435_x.jpg
    P2360435_x.jpg
    757.6 KB · Views: 922
  • P2360438_x.jpg
    P2360438_x.jpg
    644 KB · Views: 924
Yes, and it shifts as you move across the board. Lower right is worse case, upper left is OK. The next board will be different from this one. You can find several other issues if you look close enough.

This is a DIY proto board that was made using an iron on toner transfer method. My printer can't make two identical prints, and the paper itself goes through thermal shock when the hot iron hits it. These protos will have registration issues between the two sides and well as several other less than desirable traits like lack of plated through holes. They will never leave my lab, so I live with the drawbacks in order to eliminate the time and $$$$ for one or two off boards.

It is the fastest way to get a board made so that I can verify the layout and test various layout changes before dropping a kilobuck on quality made boards from a PC board house.

I do several of these over the course of designing something. When done, they may get saved, stripped for parts, or used for experiments. I went through maybe 6 or 8 board spins for the original TSE design since the amp design itself was new and still in a state of flux, as was the PCB layout. In this case there were minimal changes to the amp design itself, so there should be only one or two proto board spins. I will know more once this one sees power up and torture testing.

The TSE-II is pretty much the SAME as the original TSE, yet DIFFERENT.

The board is larger to reduce overcrowding with 300B's and fat 2A3's.

The parts list is being redone to use common parts that are available from multiple sources. Several different flavors of electrolytics and mosfets have been ordered for testing.

Unlike the original TSE, component price is not a selection criteria. The price difference between budget parts and good (but not stupid) parts is nothing in the overall amp cost.

The only change to the audio path was to add gate stoppers to the mosfets. Mosfets in general have become much faster than the old 2SK2700 used 15 years ago. The gate resistor adds some insurance against oscillation.

The filament supply has a new regulator, 3 input filter caps in parallel instead of one big one to lower overall ESR and improve ripple current capability. The goal was to make operation with 5 amps worth of 2A3's more reliable.

The layout was changed to improve separation between the filament current loops and the audio path ground point.

A zener diode / resistor pair was added to the mosfet positive supply to reduce mosfet power dissipation, and each TO-220 device now have their own standard size heat sink.

The parts layout is pretty much the same as before but things were moved around a bit to avoid thermal hot spots that were present in the old board when operated at high voltages.

This board (and all that follow) will be tested and optimized at B+ voltages from 250 volts to over 400 volts and a chart for resistor values at each voltage will be generated. All power supply caps are rated for 105 degrees C, 450 volts and chosen for ESR, ripple current capability, and lifetime.

The screw connectors as used in the SSE and SPP are now used in the TSE-II. Connections were added for a supplemental cap and choke, as in the SSE.
 
Last edited:
George,

Good luck with your experiments!

Looking at the proto PCB just remembered another aspect that can be (maybe) improved - please refer to the the MOSFETs and heat-sinks within the blue rectangle of the attached picture. In my experience, this place was a tight fit for the sinks and the height was a challenge - if your sinks are 1" tall, your PCB has to be mounted lower from the top, and the 5842 tubes become nearly invisible. Other option is to keep the heat-sinks exposed, but that might pose electrical hazards and might not be aesthetically acceptable for some folks.
 

Attachments

  • P2360442_x.jpg
    P2360442_x.jpg
    370.3 KB · Views: 803
Last edited: