TO-39 Can substitute

Thread Starter

cmstratton

Joined Jan 7, 2016
3
I have an application that uses a 2N3440 transistor, which is in a TO-39b package, in a very demanding environment, high temp 300 DEG F plus, and a lot of vibration. Eventually one of the leads will break. I've tried mounting it flush to the PCB, away from the PCB, buffered with silicone, etc, but eventually it will break. Does anyone know a substitute in a flat package like a TO-220? Thank you in advance



Mods edit:
Please don't do double or triple post.
 

RichardO

Joined May 4, 2013
2,270
The MJE340 might work. The biggest problem is that it is in a plastic packge that limits the operating temperature to less than 150 deg. C. A TO-220 packaged device would have similar temperature limits.
 

crutschow

Joined Mar 14, 2008
34,285
Staking it to the board flush with epoxy should prevent any lead breakage from vibration.

Edit: The silicone you used likely had too much give and allowed the transistor to move under vibration, causing the lead breakage.
 
Last edited:

Thread Starter

cmstratton

Joined Jan 7, 2016
3
300F is 149C. TIP47, 48, 49, 50 with 2 degrees F to spare. What gain do you need?

ak
Hi-Thanks. Not sure of the gain. here is a link to the circuit http://www.logwell.com/schematics/10/CCLSIEa.pdf Note: The circuit shows two different transistors, but I use the 2N3440 in other devices, and I have a lot of them on hand, and it works just fine

Check out the name of the site: Analog Services, Inc. Why are you called the AnalogKid? These are tools that are lowered into oil wells, BTW
 

GopherT

Joined Nov 23, 2012
8,009
Hi-Thanks. Not sure of the gain. here is a link to the circuit http://www.logwell.com/schematics/10/CCLSIEa.pdf Note: The circuit shows two different transistors, but I use the 2N3440 in other devices, and I have a lot of them on hand, and it works just fine

Check out the name of the site: Analog Services, Inc. Why are you called the AnalogKid? These are tools that are lowered into oil wells, BTW
Few silicones have any structural strength at your operating temp. They will flex like the silicone is not even there. Use a high temp epoxy - contact masterbond.com for a recommendation.
 

SLK001

Joined Nov 29, 2011
1,549
I used to work in the design of Dual Laterallog down hole logging equipment. I am familiar with the temperature requirements. However, where is the vibration coming from? How is the device failing? Are the leads fracturing? Is the device breaking at the solder joint? Are you using high-temp tin-silver solder?

Our tools precluded the use of any plastic package, as we had a design spec of 325 F. We never experienced any failures like this.
 

Thread Starter

cmstratton

Joined Jan 7, 2016
3
Thank you for all of the input. These tools are lowered into water injection wells, and consequently there is a lot of lateral movement. The break usually occurs where the collector is welded to the can. I think that the mass of the can is too much for one lead, and consequently it breaks. Thanks again.
 

SLK001

Joined Nov 29, 2011
1,549
There is (at least used to be) a heatsink that mounted via a stud on a PC board in which a TO-39 can is inserted. The leads are then bent over into the PCB holes. If these don't exist any more, then have a hole (the diameter of the can)/pad made in the PCB, then mount the can into the hole and solder it to the board. Then bend the leads to their holes.
 

crutschow

Joined Mar 14, 2008
34,285
I agree with Peter.
Such a glue mounting procedure was also used on the aerospace boards I worked on that had to survive high levels of vibration.
Staking the transistor with the proper glue should prevent the failures you've noted.
If the transistor can't move, then vibration can't cause lead flexing and fatigue failure.
No need to change the transistor or add some more complex mounting scheme.
 

AnalogKid

Joined Aug 1, 2013
10,988
If the transistor can't move, then vibration can't cause lead flexing and fatigue failure.
Not always true. MIL-STD-810 and MIL-STD-901 have some extreme test requirements, and I have seen individual transistor leads and film and ceramic capacitor leads break, and relatively small aluminum electrolytics with a 360 degree continuous bead of RTV come flying off the board. For some magic reason, 1/4 W metal film resistors seem to be immune.

ak
 

GopherT

Joined Nov 23, 2012
8,009
And the tin can on this transistor likely has a simple butt joint to the brazed/soldered collector pin. I doubt there is any mechanical attachment (e.g. tapered pin) to the assembly.
 
Top