The ultimate DIY? A Stax SRM-T2!

[johnwmclean]

How critical are you guys with heatsink surface smoothness? As a standard I wet hand sand finishing with 1200 grit paper then buff to a shine, trying my best to be as even and flat as my tools allow.

 

[Craig Sawyers]

Depends.  I used big Fischer heatsinks on my power amps, and the mounting face was pretty banana shaped - actually fairly typical of big section extrusions.  Took quite a bit of work to level them (all 8!) - abrasive paper taped to a glass plate.  But in fairness to Fisher, they say that if surface flatness is important they should be machined flat (which they offer to do - for an undisclosed price).

But polishing isn't really necessary - heat sink goop fills in minor surface irregularity.

But I did not flatten my T2 heatsinks, and have not had any problems.

[headinclouds]

4 hours ago, Craig Sawyers said:

Depends.  I used big Fischer heatsinks on my power amps, and the mounting face was pretty banana shaped - actually fairly typical of big section extrusions.  Took quite a bit of work to level them (all 8!) - abrasive paper taped to a glass plate.  But in fairness to Fisher, they say that if surface flatness is important they should be machined flat (which they offer to do - for an undisclosed price).

But polishing isn't really necessary - heat sink goop fills in minor surface irregularity.

But I did not flatten my T2 heatsinks, and have not had any problems.

Very true - they certainly are banana shaped.  When I made my first cases I decided to bolt a 1inch square bar (bit OTT) to the heatsink and tap that vertically downwards to hold the transistors.

When I got the heatsinks and bar neither was flat enough so I machined a quick skim over the heatsink and the bar to get flat mating surfaces.  But even having access to a milling machine doesn't totally solve the problem if the parts are not clamped correctly they'll bend and still not be flat after machining.

P.S John, your fellow Australian has some useful info on all this in case you hadn't come across it.   http://sound.westhost.com/heatsinks.htm

Edited April 24, 2016 by headinclouds
add PS

[Kerry]

The Conrad heat stinks seem to be ground flat and I have not had any issues. In my case since I'm mounting the transistors directly to the sinks, if they are bowed the ceramic isolators would crack when you torque them.

I don't do any additional treatment to them.

[wokstarr26]

Agree with Kerry, Conrad heatsinks are dead flat well made. Only issues it is so hard to buy it from them. My last purchase took weeks. I am in Aust dealing directly with them. In June I am traveling to Melbourne I may just drop into their factory outlet. If any one wants some heatsinks then I can buy it over the counter and ship it myself.

[johnwmclean]

Thanks for the link Geoff.

I agree the Conrad sinks are finished very nicely, for my Carbon I’m using a Chinese supplier and the finish is not so great as it has a deep brushed finish. I’ve had the sinks machined by Cam-Expert and had cavities milled for the Cree devices, which I’ve also sanded and polished as the drag lines from the cnc were quite rough.

Anyways here’s a shot of a WIP.

IMG_6399 

 

[Kerry]

I'm happy to report that I solved the oscillation issue on the right channel.  Things are looking good.

Still need to solve for the DC issue in the left channel.  The -543 TP is sitting at -507 with the top of both batteries sitting at +233V.  The outputs sit properly with the same adjustment on the 510R for the left and right channels.  I'm going to change the collars and isolators, just in case there is some leakage there.  It's curious that it's pulling both legs up equally. 

Thoughts?

EDIT:  I've thought of doing some similar milling of heat sinks John.  Looks like a viable solution.

Edited April 25, 2016 by Kerry

[spritzer]

I had the same issue and I'm pretty sure it was simply leaking to ground somewhere.  At these voltage levels it is to be expected. 

[JoaMat]

1 hour ago, Kerry said:

I'm happy to report that I solved the oscillation issue on the right channel.  Things are looking good.

Good news. What was the problem and its solution? Might be valuable information for us.

[Kerry]

I increased the 5pF bypass cap across the 100k feedback resistors to 15pF. I only need this in the right channel.

I'm going to work on my DC issue tonight.

[JoaMat]

Glad to hear it was that easy get rid of the oscillation.

My idea is to build a version with ksa1220/ksc2690 replacing the old and hard to get mosfets. Have listen to an original DIY T2 with the mosfets replaced for a week. My impression is that it works well.

[Kerry]

Craig gave me the idea to start there when he was talking about oscillation due to the feedback lines.  I'm not sure if mine is the source of the problem or if I'm just masking something else, but it works perfectly.  The feedback paths are fairly symmetrical between the two channels and fairly short runs compared to the original boards, so not sure why this was happening.

[Kerry]

14 hours ago, JoaMat said:

Glad to hear it was that easy get rid of the oscillation.

My idea is to build a version with ksa1220/ksc2690 replacing the old and hard to get mosfets. Have listen to an original DIY T2 with the mosfets replaced for a week. My impression is that it works well.

This is great and gives options for using all current/available parts.

[sorenb]

@kerry: if some wants to build the DIYT2 based on your layout will you provide the latest gerbers?

 

[Kerry]

Still working to make sure it's stable.  I'm very close, just working on the DC issue on the left channel.

I went back and replaced Q16 - 20 on the front right battery because the noise was still present, thought I could fiddle with the adjustments to make it go away.  The sand replacement has completely fixed this.

I'm having an issue where I can't stop listening long enough to look at the electronics

[iwik]

Guys,

I have completed my power supply and all went well. Now ready to assemble the Amp in the case. I have a dilemma, holes in the pcb for the

octals are 1.9mm and my sockets have a bump on the pins which are 2.3mm. Can i drill out the holes or do i have to file this bump off

What was the reasoning behind putting the pots on the backside ?. As i may have to change my pots to the other side due to miscalc when

i made my case. My amp pcb is the original version from Kevin. All comments appreciated

Thanks 

[kevin gilmore]

you don't want the tube sockets all the way in otherwise they are too low for the chassis.

[GeorgeP]

47 minutes ago, iwik said:

Guys,

I have completed my power supply and all went well. Now ready to assemble the Amp in the case. I have a dilemma, holes in the pcb for the

octals are 1.9mm and my sockets have a bump on the pins which are 2.3mm. Can i drill out the holes or do i have to file this bump off

What was the reasoning behind putting the pots on the backside ?. As i may have to change my pots to the other side due to miscalc when

i made my case. My amp pcb is the original version from Kevin. All comments appreciated

Thanks 

Agree with Kevin, but I think he is referring to original his chassis design.

I am guessing from what you are saying that with your top plate installed that there is no room for the trimpots to be on the up facing side of the board - is that right? You don't "have" to mount the trimpots on that side, it just makes it easier to adjust the batteries with the bottom plate on and the amp sitting flat.

And I am also guessing that your concern with the octals is that they would protrude too far above your top plate? You could go either way I suppose, just bearing in mind the you if you drill the board and all the other parts are installed, you run the risk of some conductive debris being lodged somewhere unknown until the time that you power on the amp. or you could check ebay for some sockets that don't have the bumps in the pins.

 

 

[iwik]

Thanks guys,

Kevin  in my chassis i have to have the octals  well down to threaded portion and the 9 pins bearly thru the board.

George  i think my pots will be fine after a remeasure. Re the cuttings from redrilling, as i have all my parts elevated i think i hopefully

can keep everything clear. Did your octals have a bump half way down the pin?. Will take your suggestion and go hunting.

[GeorgeP]

Yes they did (they were the CMC ones), but the spacing was perfect for my purposes. Down to the threaded portion seems odd though - wouldn't that mean the actual socket portion is sitting only a few millimetres from the board?

[iwik]

George,

Yes you would be right. I see the original statement from Kevin was to put Octals down fully and then the 9 pin ones just barely

thru the board, is that what you did. May just do that and if my sockets sit a bit to high then may disguise this with  rings surrounding the base.

As a matter of interest what was the height diff between sockets on yours

Thanks

Les

Edited April 30, 2016 by iwik

[GeorgeP]

It would look like this:

 

[iwik]

Thanks,

Still a bit hard to see exactly how much higher one is than the other. Any way i just soldered the octals hard down and the

9 pin with the pins just protruding thru the other side of the board.

Les

[johnwmclean]

On 4/24/2016 at 9:08 PM, headinclouds said:

P.S John, your fellow Australian has some useful info on all this in case you hadn't come across it.   http://sound.westhost.com/heatsinks.htm

Reading this article again... 

“I wonder how many readers have come across (especially) TO-220 transistors that look as if they had been machined with a chainsaw. A flat file, or fine wet/dry abrasive paper and water (no, it won't hurt the transistor) should be used to make sure that the surface is as flat and smooth as you can make it. In many instances, a device with poor finishing of the metal face may be indicative of a fake - counterfeit transistors are almost guaranteed if you buy at a very low price from any online auction site.

Never hold the transistor in a vice and file it!  Hold the file still, and gently slide the transistor on the file (or wet/dry paper on a sheet of glass) until it is smooth. There is no point having it nice and smooth if it is rounded - this will happen if you rigidly mount the transistor and hand-hold the file.

When you are finished, you should be able to lay the transistor on the heatsink and see no light between the two surfaces, from any observation angle. Use a bright light behind the heatsink so you can see any surface imperfections. This little bit of extra effort may mean the difference between the success or failure of your project - at least in the long term.”

I’ve got some mildly roughed up 10M90s on the mounting tab surface area, the al oxide pads I imagine aren’t as forgiving as a mica pads in terms flattening out in air pockets and filling in surface imperfections. It has me got me thinking or am on OCD overdrive.

[johnwmclean]

^ Here is a before and after shot from trying the above, worked a treat, I’m sure there will be some degree of thermal benefit.

IMG_4947 s