DIY mini T2 Build Thread

[JoaMat]

I like the idea about to-252.

First draft… 

 …have to think about this.

XLR ground pin to signal ground on PCB.

 

 

 

[Shawn]

12 hours ago, JoaMat said:

XLR ground pin to signal ground on PCB.

Correct me if I am wrong.

The Ground Pin(Pin1) has been connected to the XLR socket shell by default by the manufacturer in some cases.

This way, the ground pin should be left unconnected to avoid a looping ground.

In this case, the loop may be like this: XLR(Pin 3)-> AMP signal ground - > L bracket - > chassis -> XLR(Pin 3).

I just realised the L-bracket is isolated to the signal ground, so it should be fine.🤐

I assumed the amp chassis ground wires are connected to the PSU chassis somewhere, and the amp signal grounds are merged with the XLR ground, and then connected to the PSU power supply ground somewhere.

[kevin gilmore]

the neutrik jacks i used in the diy t2 are actually 4 wires. the connector shell is a seperate wire. i put in a jumper to chassis ground for that wire.

[JoaMat]

You might say I follow the strategy of DIY T2 PSU by Kevin. I have two PSU I’m using for all my electrostatic amplifiers. One original T2 PSU and one I made myself basically same as the original but with KGSSHV PSU regulators.

Bottom line: I’ve never have bother about grounding strategies. So far, I haven’t had any issues with ground loops that I’m aware of.

[JoaMat]

On 4/27/2025 at 11:20 AM, Shawn said:

IXTP10M90S can be replaced with IXCP10M90S(TO-252). If KSC2690 can be replaced by something else with an SMD package, we can eliminate all the aluminum oxide isolations.

Milled a small board to test IXCP10M90S(TO-252).

Attached board to heat sink via L-bracket and sil pad for insulation purpose of course. When increasing current to 20 mA left side its 10m90s popped and left the board completely. Found it on floor.

So, no more testing regarding surface mounted 10m90s.

I’m very pleased with the new mini T2 board.

Here with 300B via tube adapter with filament injection, 5VAC from transformer. Works great. Nice setup, don’t you think?

[Shawn]

32 minutes ago, JoaMat said:

Attached board to heat sink via L-bracket and sil pad for insulation purpose of course. When increasing current to 20 mA left side its 10m90s popped and left the board completely. Found it on floor.

That 10M90S decided it was done with this world and took a leap of faith! Looks like it couldn’t handle the heat.☹️

Did you check this section? Looks like it was shortened by the solder.

[JoaMat]

Thanks for notifying. The more eyes the better and… No, I missed that damage.

With L-bracket it looks like this.

Clearence between +400V area and ground is 40 mil. So, I doubt that there has been an electric short. I suspect that, for a reason I don’t understand, the current rushed and 10m90 quickly over heated. At the same time the PSU gave up and all went silent. I guess it’s heat from 10m90s that caused the damage. But that is a 100% guess.

Anyhow, I think I stick to to220 version of 10m90s and forget this mishap.

[simmconn]

Yep, if you had checked the theata-jc of those packages you wouldn’t have bothered with such “test”.

[JoaMat]

Since I don’t know how to calculate the thermal resistance between to-252 and heat sink out into free air I made this small test. I got a clear answer. I do know Ohm's law though.

[simmconn]

Not exactly the same part but being the same package they are more in common that they are different:

https://fscdn.rohm.com/en/products/databook/applinote/ic/power/linear_regulator/to252_thermal_resistance_information_an-e.pdf

In the final stage CCS, the 10M90S is dissipating 8W to 9W. The theta-JA needs to get down to single digit °C/W in order for the junction to stay comfortably within spec, not to mention that the output DC offset thermal drift has always been our enemy. TO-252 is not going to get you there. Also, even if the copper is well coupled with the aluminum angle, the tiny cross section would still give considerable thermal resistance. 

You are right in that it is just like the Ohm's law. With theta-JC being 3.1 already, there is not a lot of slack to play with. I think IXYS is being sloppy when it comes to thermal specs in the 10M90S datasheet, by specifying only one number for both packages. With some other parts having about 2x theta-JC in TO-252 than in TO-220, I doubt IXYS can do that much better.

[Shawn]

14 hours ago, JoaMat said:

Anyhow, I think I stick to to220 version of 10m90s and forget this mishap.

The TO-220 version of the 10M90S is readily available, so there's really no need to consider alternatives for now. Better to stick with what’s proven to work.

[JoaMat]

Thanks for comment and link to ROHM document, @simmconn. Very much appreciated.

That test had obviously no chance to be successful.