LM3886TF Heat Sink Insulation

[kscharf]

Originally posted by jackinnj
I use these on the chipamps -- easy to make with a round file -- easier still if you have a mill:

Isn't that machined to fit the LM4780?

[jackinnj]

Originally posted by kscharf


Isn't that machined to fit the LM4780?

Yes -- I used it in the article -- pulls the device nice and snug to the heatsink -- easily machined with an end-mill and scrap aluminum. I don't have a pic of the one I used on my inverted LM3875 amp.

Here's the mounting methodology on a plate amplifier I found at the town dump:

[niacinj]

Hello,
I've pretty much decided on using the described bar method for mounting my lm3886's to the heat sink. However, I have BrianGT's board which I've noticed doesn't allow much room for the mounting bar. My question is can I solder the lm3886 on the topside (normal) and mount all of the capacitors and resistors on the underside? Or maybe just the taller caps? I can't think of any drawback to this option other than the components might need to disperse some heat?

Any thoughts?

[Andrew Eckhardt]

There is really no call for a bar on the TF package. The resistance of the molding over the back of the device is already high enough that high ands distributed force isn't going to get you anywhere important. A screw through the tab with an inch pound or two and a thin but complete film of silicone/zinc oxide paste and you're ready to rock. A little math can easily be used in place of my opinion. Always use some kind of paste and reject the rubber washers for best performance. Plastic screws are okay to load down bars, but they will stretch at normal semiconductor temperatures and possibly cause an unnecessary failure if used directly. Because of the increased resistance to sink on encapsulated packages the problem is actually worsened.

[niacinj]

Shoot! I was too busy reading about the mounting bar that I must have missed the fact that this thread was originally for the TF package. I will be using the un-insulated package with a pad for my amplifiers. I'm pretty much sold on using the mounting bar with the un-insulated chip, so my question still stands on whether I can mount the rest of the circuit to the underside of the board.

I can re-post in the appropriate thread if I need to.

[Redshift187]

Sure you can put components on the bottom of the board niacinj. The only thing on the board that really generates any amount of heat is the LM3886.

[jack86]

There are two OP of this IC, one is 3886t, and the other is 3886tf, 3886t is a metal pack one, it should be insulated from heat sink, and for the 3886tf, there is no need. want to reduce the temprature when it working, you'd better use conduction paste!

[AndrewT]

Andrew E and Jack allude to the TF package as not having a heat dissipation issue, because the TF is electrically isolated.

That is quite misleading.
I believe that the electrical isolation of the TF package makes heat dissipation a big issue. Particularly in trying to keep the chip operating temperatures low enough to get good performance from this protected chip.

A clamp bar, when done properly, is repeatedly shown as giving better heat transfer from chip to sink (Rth c-s).
A bolt can rarely if ever perform as well as a clamp bar in this respect.
The TF needs all the help you can give it to help keep operating temperatures down.

[AndrewT]

Originally Posted by niacinj

my question still stands on whether I can mount the rest of the circuit to the underside of the board.

if the PCB is double sided with Plated Through Holes (PTH) and the pads on the top side are big enough to effectively solder to, then mounting the components on the bottom side is not an electrical problem.
But be very careful that all components are mounted in the correct holes. Are there any three (or more) pin components? These may all have to be mounted on the top side unless you can bend/re-arrange to pins to fit the wrong side.

[Andrew Eckhardt]

[QUOTE=AndrewT;2290186]Andrew E and Jack allude to the TF package as not having a heat dissipation issue, because the TF is electrically isolated.

It might be misleading because the allusion doesn't happen. All I said is that high force isn't going to drop the thermal resistance of the case. You get about 2 degrees per watt out of the epoxy coated case. If you get .15 additional out of some paste with 1 inch pound on a tab mounted screw, and .12 out of a clamping bar with 5 inch pounds on each screw, what happens to the 2 degrees C of the plastic?

If you're that concerned about reliability and thermal resistance, you're definitely better off with the metal tab, and almost better off buying more of them, since they don't cost much more than a few manually drilled holes in a piece of metal.