good transistor tester in the <300$ range

[coppercone2]

are the heat sinks on the back acutally live or do you mean the transistors body?

The insulator is easy enough, but isolating a hot transistor is more difficult. I have little white covers from some old Digimatic or some thing calibrator. It had this shit, and the answer was little plastic covers over the transistors. I think they were a bit smaller then TO-3 though? (that unit had the switches like 'welded' to the PCB and when traces started delaminating i threw it out, and there was green shit growing through all the wires, and the chassis looked super burned up, I think batteries leaked on top of it a long time ago or something, I thought the PCB might be compromised).

However, if you use those white covers, they can hide a broken solder lug. It had a solder lug that was bolted to transistor body with the white cover on top. The lug fractured, but it got pinned down under the white cover, so there was a broken circuit but it looked like the solder joint was intact! Odd to see a lug fracture like that too on the neck between the hole and the solder joint. I wonder if the cover had something to do with it, but its hard to imagine why pressing it with plastic would make it break there.

The white cover should have a brass insert in them, embedded, so you just screw it on from the back and it insulates the whole transistor. It also might have a few little air holes for some reason.

In any case, I would advise you to clamp the lug through the bolt on the back side with some spacers, instead of clamping it between the transistor and the cover, because while I can't explain WHY, I have a bad feeling the lugs don't like being clamped by that plastic. maybe it got too hot and moved around too much and stressed it some how.


This looks similar
https://www.newark.com/keystone/4632/power-transistor-cover/dp/97H7266

https://www.cariatielettronica.eu/gb/other-series/5769-2-pezzi-cover-case-calotta-di-protezione-per-transistor-in-formato-to3.html

[coppercone2]

it identifies a NIR laser as a LED

[factory]

are the heat sinks on the back acutally live or do you mean the transistors body?

The insulator is easy enough, but isolating a hot transistor is more difficult. I have little white covers from some old Digimatic or some thing calibrator. It had this shit, and the answer was little plastic covers over the transistors. I think they were a bit smaller then TO-3 though? (that unit had the switches like 'welded' to the PCB and when traces started delaminating i threw it out, and there was green shit growing through all the wires, and the chassis looked super burned up, I think batteries leaked on top of it a long time ago or something, I thought the PCB might be compromised).

However, if you use those white covers, they can hide a broken solder lug. It had a solder lug that was bolted to transistor body with the white cover on top. The lug fractured, but it got pinned down under the white cover, so there was a broken circuit but it looked like the solder joint was intact! Odd to see a lug fracture like that too on the neck between the hole and the solder joint. I wonder if the cover had something to do with it, but its hard to imagine why pressing it with plastic would make it break there.

The white cover should have a brass insert in them, embedded, so you just screw it on from the back and it insulates the whole transistor. It also might have a few little air holes for some reason.

In any case, I would advise you to clamp the lug through the bolt on the back side with some spacers, instead of clamping it between the transistor and the cover, because while I can't explain WHY, I have a bad feeling the lugs don't like being clamped by that plastic. maybe it got too hot and moved around too much and stressed it some how.


This looks similar
https://www.newark.com/keystone/4632/power-transistor-cover/dp/97H7266

https://www.cariatielettronica.eu/gb/other-series/5769-2-pezzi-cover-case-calotta-di-protezione-per-transistor-in-formato-to3.html

Yes those three heatsinks are live  , along with the transistor cases, given that this is a 300V 100mA current source and those are the regulator transistors this is not good.
They will get relocated back inside the PSU when I've sourced some ceramic TO-3 insulator, as some previous owner lost the boron nitride insulators. Probably need to rewire that part when I move them, as it's quite a mess.

These are what I've found so far.
https://www.ebay.com/itm/233510502749

David

[Gyro]

it identifies a NIR laser as a LED

A laser diode is an LED until it reaches the 'LASEing' current threshold.

[factory]

it identifies a NIR laser as a LED

A laser diode is an LED until it reaches the 'LASEing' current threshold.

It can also identify stabistors as LEDs, for obvious reasons.

David

[David Hess]

Syscomp Electronic Design makes a 1 amp 30 volt USB curve tracer if you can figure out how to buy it:

https://www.syscompdesign.com/product/ctr-201/

[coppercone2]

are the heat sinks on the back acutally live or do you mean the transistors body?

The insulator is easy enough, but isolating a hot transistor is more difficult. I have little white covers from some old Digimatic or some thing calibrator. It had this shit, and the answer was little plastic covers over the transistors. I think they were a bit smaller then TO-3 though? (that unit had the switches like 'welded' to the PCB and when traces started delaminating i threw it out, and there was green shit growing through all the wires, and the chassis looked super burned up, I think batteries leaked on top of it a long time ago or something, I thought the PCB might be compromised).

However, if you use those white covers, they can hide a broken solder lug. It had a solder lug that was bolted to transistor body with the white cover on top. The lug fractured, but it got pinned down under the white cover, so there was a broken circuit but it looked like the solder joint was intact! Odd to see a lug fracture like that too on the neck between the hole and the solder joint. I wonder if the cover had something to do with it, but its hard to imagine why pressing it with plastic would make it break there.

The white cover should have a brass insert in them, embedded, so you just screw it on from the back and it insulates the whole transistor. It also might have a few little air holes for some reason.

In any case, I would advise you to clamp the lug through the bolt on the back side with some spacers, instead of clamping it between the transistor and the cover, because while I can't explain WHY, I have a bad feeling the lugs don't like being clamped by that plastic. maybe it got too hot and moved around too much and stressed it some how.


This looks similar
https://www.newark.com/keystone/4632/power-transistor-cover/dp/97H7266

https://www.cariatielettronica.eu/gb/other-series/5769-2-pezzi-cover-case-calotta-di-protezione-per-transistor-in-formato-to3.html

Yes those three heatsinks are live  , along with the transistor cases, given that this is a 300V 100mA current source and those are the regulator transistors this is not good.
They will get relocated back inside the PSU when I've sourced some ceramic TO-3 insulator, as some previous owner lost the boron nitride insulators. Probably need to rewire that part when I move them, as it's quite a mess.

These are what I've found so far.
https://www.ebay.com/itm/233510502749

David

Lol for some transsitor those ceramic heat sink are shenanigans. The boron nitride ones I had on the 6177A disintegrated. The ones on the 6181C #1 and 6181C #2 crumbled a few edges but they were mostly OK. So I was happy to reuse them.


However, I did do all the repair stuff on the 6177A where the boron pad disintegrated and replaced it with alumina pad. It did not fit in the square hole. Alot of the alumina pads you can get are too big. I had to trim them with a diamond wheel dremel after marking them in the shop.


I also previously run into problems where alumina insulators were too thick to fit in HP stuff, so I had to replace all the bolts, but in your case the bolts look like they should be thrown away. And in another case, they were too thick, and the part leads were bearly accessible. I forgot where that was. Its a problem when its PCB mount and not wire mount. But that happens when you replace mica with alumina.


I think it might be possible to make a boron nitride pad, if you have a semiconductor silicon saw and some money
https://www.amazon.com/Heloawei-Ceramic-Rods-Nitride-Diameter/dp/B0BWNLLQ7V?th=1

I got really pissed, because its a nice material. I am sure you can get it cut at a higher end machine shop because its dimensionally simple. It might not be such a advanced operation as we think, as long as they regularly make similar stuff. But I have a feeling they usually use heat to make these, instead of machining them, but it might be a option.


If you are not familiar with it, the boron nitride pad is actually ROUND, and they have a fiberglass 'cutout' that goes around it to match the transistor outline. So all you really need is the right shape BN round disk with 2 holes, and a custom PCB cut out. The BN is not the same shape as the alumina pad, which is all inclusive




I don't know why, but I suspect that someone did not lose it, it actually disintegrated. Mine turned into like 10 pieces as soon as I undid the screws. I thougth maybe they tighten it good in factory, and the aluminum casting has a memory / mind, and over the years it warp and smash the insulator. I heard some castings behave strangely.

you might want to check out that heat sink with some dye to see if its flat too. so you don't break more pads

[factory]

That explains a lot, I was wondering why the picture from the 6186C had PCB type material showing, the ceramic pad being round explains that. There is some flexible PCB mixed in there too, that is long gone on mine. The PCB on the other side is only found in the 6186C, it contains a bunch of Zeners.



This is what this 6186B is like inside, the surface of the heatsink will need work, as there is some corrosion, it's a pity there is a hole that the wiring is going through, that shouldn't be there, it's more work of the 'phantom tape bomber' responsible for the other bodgery.





According to fischerelektronik, these aluminium oxide insulators cut be laser cut, not sure normal machining would be a good idea, it's too easy to break.

There is another grade on boron nitride that is much harder to break (grey color), used for the lathe tips at work, a 5mm cut will shatter them eventually.  Those 12mm diameter tips are approx £300 each.
No idea how much amazon are pricing the regular grade at, as the geo-blocking is active for the link you added. 

No one seems to make the round TO-3 pads. Originals were HP #0340-0411, boron nitride part #H-4001, made by Union Carbide Corp.



https://www.fischerelektronik.de/fileadmin/fischertemplates/download/Katalog/heatsinks.pdf

https://www.mouser.com/catalog/specsheets/Aavid_Board-Level-Heatsinks_Cat.pdf

David

[coppercone2]

Its hexagonal pressed BN right? not CBN? One is ALOT softer. They could think you mean CBN. If its CBN then you probobly need lasers... its going to work like complete shit

https://www.bn.saint-gobain.com/products/machinable-ceramics#

I have seen the darker color BN pads in the shape of the Alumina pads I think, in more serious equipment (navyish)

EH I think you can cut it with diamond if you have a big wheel and its cooled well. Its just a massive pain in the ass and probobly pretty expensive to cut compared to a laser. Those places are VOLUME. No one likes it when a $30 tool makes 1 cut. There are like 4 different types of diamonds. You can get fully sintered (solid diamond bearing material), plated, plated and layered or brazed. (and all that in CBN too) And they have single crystal diamond tools and polycrstaline diamond tools for lathes (the single crystal looks like a ring diamond, and those can be low quality yellowish or high quality clear). And there is probobly more options. Its a bitch to find these diamond cutters... alot of people don't know they exist. Best bet would be to find one of the better ones and show it to someone, because likely they are thinking "there is no way a 2 layer plated diamond can do this" (cheap ass tile saw type thing) but it might not be the case for a 5 layer plate or a diamond matrix/composite (engineering cutter for $100)

The heat sink will cleanup some what if you ultrasonic it with alnox from experience. The area under the transistors is usually much better quality (fly cut maybe) then the back side. Usually its painted on the air heat exchange side. If its actually nice and you want to clean the rest, put some tape over that to protect it from the chemical action for the long duration cleaning, you can clean it for a short period of time without the tape at the end before washing.


Maybe this bad boy, if you can get it in like 1/8 inch
https://www.hisglassworks.com/6-inch-x-3-8-inch-100-grit-sintered-diamond-straight-edge-wheel.html
10mm diamond depth


Most people think of 'plated' diamond (microns) not milimeters.



If you have the stomach to find it here
https://ukam.com/diamond-cbn-wheels-with-smart-cut-technology/


then you would need to lap it on diamonds.


If their not perfectly uniform between cuts, who cares, just mill the PCB to make it a little thinner. transistor to transistor variation in heat spreader thickness is irrelevant, it won't do shit if its 0.01 inches spread between 4 transistors. The PCB is easy to adjust in thickness. They just need to be flat and close. It would probobly mean lapping.




But honestly, lasers are getting common. Maybe its alot cheaper to not do it yourself.

[coppercone2]

but the alumina are not hard to cut, I did them free hand with a dremel.


YOu clamp it to the heat sink  with some washers without the transistor, and just cut along the heat sink to make it fit 

I the first one break a little before I got a hang of it. But otherwise they cut nice and smooth, just had a little bit of a goofy appearance (I used a dremel brand diamond wheel)