Heat sink question

[leydenjar]

I have an mc7812c voltage regulator in a boogie express that I'm repairing that doesn't have it's heat sink mounted to anything. It's terminals are connected in the circuit but the package isn't mounted. Can I safely mount it to the chasssis?

I Forgot to mention it's in the heater rectification circuit. I'm not familiar with this strategy at all. If anyone has time to explain what it does or point me to an article/post about this implementation I'd be very appreciative.

[Chuck H]

Never seen that. Can't imagine the motivation. Unless that particular part never dissapates so much as to actually need the heat sink properties and Mesa figured to save the labor of mounting it. If it's not the problem "I" would just leave it alone and repair whatever is broken in the amp. If your REAL interested you can call Mesa. They're usually very helpful.

[g1]

It's got thermal paste all over it so it must have been mounted to something. Is there no evidence of thermal paste/grease anywhere on the chassis or elsewhere that it could have been mounted to? If it was just meant to be hanging there, why would they put thermal paste on it? At least they would cover it with heatshrink or something as the metal tab is connected to ground and could short something out.
Yes you can mount it to the chassis. The tab is ground so no insulator should be required. I think it was probably mounted to the chassis originally.
As far as the circuit goes, it's purpose is to provide regulated 12VDC for the heaters, instead of running the heaters on AC. This is not uncommon for higher gain amps where any hum reduction helps.

[leydenjar]

It's got thermal paste all over it so it must have been mounted to something. Is there no evidence of thermal paste/grease anywhere on the chassis or elsewhere that it could have been mounted to? If it was just meant to be hanging there, why would they put thermal paste on it? At least they would cover it with heatshrink or something as the metal tab is connected to ground and could short something out.
Yes you can mount it to the chassis. The tab is ground so no insulator should be required. I think it was probably mounted to the chassis originally.
As far as the circuit goes, it's purpose is to provide regulated 12VDC for the heaters, instead of running the heaters on AC. This is not uncommon for higher gain amps where any hum reduction helps.

I actually found it that way along with other strange things. I think someone attempted to repair this amp that shouldn't have. Yes there is no visible thermal paste anywhere on the chassis and the voltage regulator is mounted on a very small piece of pcb though I found it just as you can see it in the photo dangling in the chassis.

I think I will ask Mesa tomorrow.

Thanks folks.

[nickb]

I actually found it that way along with other strange things. I think someone attempted to repair this amp that shouldn't have. Yes there is no visible thermal paste anywhere on the chassis and the voltage regulator is mounted on a very small piece of pcb though I found it just as you can see it in the photo dangling in the chassis.

I think I will ask Mesa tomorrow.

Thanks folks.

The datasheet http://www.fairchildsemi.com/ds/LM/LM7805.pdf shows the heatsink is connected to the center pin 2. The schematic you supplied doesn't show that pin, but we can be pretty sure it goes to the -ve end of C50. Also the circuit is incomplete. It looks like the the ground connection to the -ve end of C50 is missing (as the diode bridge -ve is connected to gnd). This it looks a though the heatsink of the regulator and the chassis are at the same potential so you should be OK to connect it to the chassis. To be quite certain, put a voltmeter between the chassis and the heatsink on the regulator.

[Enzo]

How about this. Someone smeared heat grease on it thinking that heat grease all by itself was a cooling agent.

[Chuck H]

Is this it?

[Enzo]

Hey, that looks more like it than my idea.

[g1]

Nice catch Chuck. In leydenjar's original pic, I noticed the very long screw coming through the chassis and wondered whether it may be some kind of retrofit that displaced the original location of the regulator.
Looking at your picture, it appears to be pretty much the same location. Also there is some discolouration around the screw where the heatsink grease may have been cleaned off the chassis.
Pretty silly of someone to just leave the reg. flopping in the breeze though.
Leydenjar: what is that long screw being used for?

[trobbins]

Based on 3x0.15=0.45A at min reg margin of say 1.5V gives a min reg dissipation of 0.7W, and a free air TO-220 package has about 54C/W to ambient, so 40C junction rise - and this would be per se acceptable. However given that reg margin is more likely to be much larger, then diss is circa 6Vx0.45A=2.7W - which far exceeds junction limit and reg would enter thermal shutdown protection.

Apart from a TO-220 insulator kit, you could also use a makeshift insulator using heatshrink around the device and a spring clip pushing it on the chassis. Just adding a small uninsulated heatsink to the TO-220 may be acceptable, but then you have to restrain the heatsink from touching chassis, and junction temp rise may be marginal if ambient is high.

[Enzo]

Isn't the big long screw one of the chassis mount bolts?

[52 Bill]

Isn't the big long screw one of the chassis mount bolts?

I think that it's that grounding screw that connects the chassis with the top cabinet shield, that's the only one that I ever see in the middle of the chassis.

[leydenjar]

The long screw does indeed connect the chassis to the top cabinet shield though it doesn't screw into the top of the cabinet. It simply rests on it less than gracefully. Last minute retrofit by Mesa?

The pic Chuck H attached looks like what I intended to do with this device. There is a hole in the chassis there and it should make a good mounting spot. Apparently it was designed this way.