TO-220 'lock' washer

Jul 15, 2019 15 Replies

So we had a few transistors in to-220 pacs de-laminate at the metal tab/ plastic interface. (lost electrical connection to the collector.)

formatting link
I bent it up...it was just cracked along the edge. The circular mark around the mounting hole was made by the belleville washer. Which I think might be the problem. Washer is from Seastrom mfgr. 5809-13-16-P,
formatting link



These take a lot of force and have less defection than other belleville's. I also think they were put in with too much force.



I'm wondering if there are any spec's for lock washers on TO-220 pacs? These are being used as temperature sensors over a large range, so I do want a lot of compliance (or deflection.)



Thanks George H.


I found this, but not much info on the washer.

formatting link

GH

Wavy washers are nice.

John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com

I found a lot of stuff on the web, but mostly about getting heat out. (which I don't care about.) I think maybe stack a few bellevilles in opposition to get more defection and less force.

Do they make wavy washers small enough for to-220? (#4 or #6 screw)

GH

I follow the advice in a Motorola/ON Semi SCR/Triac manual which is to place a flat washer between the TO-220 tab and the belleville. Also be sure to use a torque driver set to the recommended torque (I forget right now which figure). Belleville is nice because it is kind of constant force.

piglet

I have seen metric stainless wavy in all sizes from M2 (2mm dia) upwards.

piglet

The first place to look for any kind of hardware what-if question is

formatting link
just put wave washer in the search box, they have #0 to #10 in 18-8 stainless steel. They also have steel and three point wave washers and belleville washers (and yes, they buy from Seastrom, I've gotten bags from mcmaster that said Seastrom on the label :-)).

Regards, Carl Ijames

The circular mark has cut right thru the plating into the tab base metal

- I think that shows the system was way over-torqued.

piglet

mark

h I

,
l

And said torque would twist the tab leading to the case being under stress, if the leads were soldering in previously to the torquing down, and eventually separating as it did.

A tiny dab of Green Locktight would probably be enough for the 4-40 nut and washer if you are mostly looking for securing with a minimum of torque...

formatting link
nd-purple-threadlockers/

John

Which is why you have to tighten a Belleville to the specified _deflection_ rather than the specified _torque_. Otherwise you're liable to bottom it out and make this sort of problem.

Cheers

Phil Hobbs

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com

[using Belleville washers]

The scraping of tin plate indicates they were installed bell-down, but not necessarily that force was excessive. More important, would be a torque scewdriver setting so as not to 'max out' the spring and keep going to higher force (hard edges made a scrape, cannot tell force from that).

Usually, one puts a spacer over the hole, and attaches through the spacer, so that the diameter of the washer doesn't impinge on the epoxy. Mechanical lateral force on a Belleville washer is HUGE, can easily crack things. The spacer, on the other hand (like a washer) spreads the steel screw pressure over the softer copper of the TO-220 tab. Loctite makes something that will retain the screw (but nickel plate screw into galvanized or aluminum doesn't play nice with the iron-catalyzed variants).

Thanks piglet, I've always used a flat washer on the 'bottom'. We'll probably add one.

George H.

Right, I think that's how I found Seastrom. (McMaster-Carr has a nice search engine... their index is a thing of beauty too!)

George H.

Yup, thanks everyone for the advice.

George H.

As a temperature sensor only, the issues with mounting of power-dissipating TO220's go away. Any lockwasher should work, if applied as designed.

Sensors may not employ the same solid-filled epoxy package structure as power semiconductors - issues with mechanical damage are different, and the requirements for environmental protection come from a completely different book.

RL

... Lock washers are useless. How many are in your car?

formatting link

Regards,

Boris Mohar

Got Knock? - see: Viatrack Printed Circuit Designs (among other things)

formatting link

void _-void-_ in the obvious place

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required