Heat Dissapation Problem LME49720/OPA1632 Preamp

[artquake]

Dear Forum,

I realized a preamp in using a schematic of alexw88 which is pictured on his homepage SCHEMATIC

I did not use a circuit board in lack of beeing able to design one . So i made it with p2p connection. I hope alexw88 will answer someday my mail so i hope to get a pair aof pcb's ?

Initial Situation:
Soldering Side:

Mounting Side:


As i have used the DGN OPA 1632 Version with PowerPad wich is soldered on an HTSSOP Adapter board, i measured max. 70°C with my IR-Thermometer. The Sound is superbe, really, alltough during hours of playing music and making measurements with RMMA it's hot like a ... whatever.

Here are some results of the measurements. I post this, beacause i find this preamp really great to be built more. As it beats in measurements the results of my older 6H30 tube preamp and also the results of my Vincent SA93 Preamp (wich in fact is a great preamp).

For example measurements with RMAA 6.06 for reference
Noise Level


Stereo crosstalk


Problem

I need a possibility to cool down the OPA1632. Maybee drilling a hole in the Adapterboard and mounting the OPA1632 vice versa, and putting some copperbar on the thermal Pad ?

Any suggestions ? Or anyone in here who can design a circuit board of this preamp ? Maybee others would be interested ...

Best regards and thanks in advance for any suggestions
artQuake

[peranders]

1 You can lower the supply voltage

2 You can clue a heatsink (of some sort) on top of the IC.

70 deg C is not a problem if you think so. If it stops at this temperature you don't have to do anything.

[artquake]

Originally posted by peranders
1 You can lower the supply voltage

2 You can clue a heatsink (of some sort) on top of the IC.

70 deg C is not a problem if you think so. If it stops at this temperature you don't have to do anything.

Voltage is allready @ 13V

Instant glue with a small massive peace of copper ?

70°C, no problem ? Really ? Also for a long period of time ? I can't find notes @ TI-Product Information
"POWER DISSIPATION AND THERMAL CONSIDERATIONS
The OPA1632 does not have thermal shutdown protection. Take care to assure that the maximum junction temperature is not exceeded. Excessive junction temperature can degrade performance or cause permanent damage. For best performance and reliability, assure that the junction temperature does not exceed +125°C."

That's why i think in my case it's critical too, because the thermal pad (very near to junction temperature) is on the downside of the ic, which on the other hand is very near to the adapter board that acts like an isolator, and as i mentioned, i measured (+/-) only the surface temperature on top of the part.

nevertheless, thanks peranders for your informations ...

[BWRX]

Originally posted by artquake
That's why i think in my case it's critical too, because the thermal pad (very near to junction temperature) is on the downside of the ic, which on the other hand is very near to the adapter board that acts like an isolator, and as i mentioned, i measured (+/-) only the surface temperature on top of the part.

That thermal pad on the bottom of the device is meant to be soldered to a copper plane that acts as a heat sink. You really should have some kind of thermal mass attached to that thermal pad. It could be something as simple/cheap as a thin strip of aluminum from a soda can that you place under the part with a little thermal paste before soldering it down. That will help much more than any heat sink you put on the top of the plastic package.

[artquake]

Any affections/influence into sound by doing this strip of aluminum or let's say copper strip to the pad ?

What do you think, when i would use D instead of DGN Model of the OPA1632 would it solve this thermal issue ? (mind you without heatsink) The problem was i currently could not find any supplier for SO-Type ...

[BWRX]

I don't know how much it would influence the sound or if you would be able to hear any difference at all. Try it and find out.

Using the SO8 package that doesn't have the thermal pad probably has worse thermal performance than the MSOP8 package with the thermal pad properly connected to some kind of heat sink (or soldered to copper on a PCB as it was meant to be).

[banana]

Take a look at the THS4631 datasheet. In which, the thermal performance difference of DGN package, with or without solder, is revealed. THS4631 cosumes similar current to OPA1632.

http://focus.ti.com/lit/ds/symlink/ths4631.pdf

Thermal resistance is 58.4degC/W for DGN with solder,
98degC/W for SOIC, and 158degC/W for DGN without solder.

I did blown one THS4631 after about an hour of operation. Since I have failed to solder the bottom powerpad to ground plate. Supply rail was +/-13V.

Driver chips for BPA300 amplifier

[artquake]

Thanks to all for the informations up to now.

The problem is that i am using this ADAPTOR, SMD, TSSOP-8, 0.65MM. And there is no PowerPad on it.

On the other way:
BWRX wrote

strip of aluminum from a soda can that you place under the part with a little thermal paste before soldering it down

Banana wrote

Since I have failed to solder the bottom powerpad to ground plate

It's somehow logic that heat transfer is better when the pad has a direct connection to a bigger materialarea or mass with a high heat capacity. How can I solder/realize this ? Is it not problematic that the IC get's to hot while soldering a bigger mass to it ? (because when heating the solder the addad mass will also heat up until the solder melts) I don't want to destroy the OPA thru heating during the soldering process.

Aren't there any MSSOP adapterboards in which a PowerPad is embedded ?

Or shall i file a gouge into the existing adapterboard and insert a copperplate into that gouge ? Like the following

I don't think thats enough area ...

As i have really pairs of "sausagefingers" for this sort of surgery job (soldering the OPA1632 on the adapter was already an extreme artifice) i would better prefer a ready done adapterboard.

Any sources available on such adapterboards in here ? Or maybe someone who's able to make some ?

Thank in advance for any suggestions ...

regards
artQuake

[peranders]

I sugegst that you glue someting on top if the IC before you do anything else.

[artquake]

OK peranders, i'll give the following SMD heat sink a try.

Rth=33K/W 6.3x30x7.5mm

Would you really glue them ? I mean the expansion of ic material during heating up/down/up and so on will lift of the heatsink with the time. In the other hand using an adhesive pad to fix heatsink will genarate more thermal resistance. Have you tried instant glue successfull with these temperatures ?

Thank you