heat sink for o2 sensor

[jcarmichael]

ok so this weekend I am installing my mtx-l wideband. putting it in the stock o2 location and doing the ecm stuff to automate my narrowband. my question is I have access to aluminum can I use that as a heat sink instead of going to buy copper and making the heat sink out of that. I tried to do a search I know alot of ppl don't run heat sinks and there have no problems but I would rather play it safe. from the searching I found which could be wrong it says copper pulls the heat out of the source quicker but transer to air slower then aluminum.

thanks for the help everyone
Josh

 

[92AWDHX40]

http://www.innovatemotorsports.com/support/manual/LC-1_Manual.pdf
copper is a great for that heat shield.

 

[jcarmichael]

yea saw thats what they suggest, ecmtuning says the same thing. just didn't know if I went with the aluminum if it was going to be a major issue or not. only say that because seeing alot of ppl not running one at all with no issues. however guess if I am going to do it might as well do it correct, or the way suggested. I really just didn't want to track down a sheet of copper.

thanks as always

 

[talonDSMerr]

You can find a sheet of copper in whatever size you need easily at mcmaster.com.

 

[jcarmichael]

theres a place here in indy I can get it from just its a big metal place and don't wanna pay an arm and a leg to get it since I'm not buying in big quanitys. rather not order online because of the waiting process, alot of ppl are running without one at all so I am going to try the aluminum because I just got my sd stuff all set up. have my daughter next two weeks as of next saturday so if im going to switch it to speed density this is my best weekend for a cpl of weeks to get my tune switched over.

 

[PieEyedPiper]

When I finally install a wb in the stock location I'm going to use hose clamps and strap my 02 sensor socket to the sensor to act as the heatsink. So many people have no issues running without one, I figure tripling the mass of the sensor, even with weak surface contact (thermal paste maybe?), ought to be better than nothing.

 

[talonDSMerr]

^ You should be careful doing it that way since most sockets are made of steel, which is a poor thermal conductor. You'd also be adding a layer of air between the sensor and socket, which is one of the best insulators, so strapping the socket on might be worse. Instead of adding mass, you want to be adding surface area, which is why sandwiching a thin plate between the sensor and o2 bung works well.

 

[Oldsid]

I would not recommend using aluminum for the heat shield. Copper thermal conductivity is about 204 Btu/(hr F ft) at 1100 deg. F while that of aluminum is about 144 Btu/(hr F ft) at 400 deg. F. That is over 40% more heat that can be removed with copper instead of aluminum. The other main reason why copper is specified is because it can be used at much higher temperatures. As matter of fact, the heat conductivity of aluminum is only given up to 400 Deg. F.
A simple and inexpensive way to make the heat shield is to use a copper tube. A good size would be an inch tube that you cut and flatten as required. You can buy cooper tube from any hardware store.
However, an important detail is to try to put the 'fin" of the heat shield in the vertical direction and make sure that no obstacle is below or above it so that heat flow by natural convection is not blocked. Sometimes this may be difficult because the O2 sensor lower flange is on the way. If you need any more details, PM and I will make sketch to illustrate what I just described.

 

[PieEyedPiper]

^ You should be careful doing it that way since most sockets are made of steel, which is a poor thermal conductor. You'd also be adding a layer of air between the sensor and socket, which is one of the best insulators, so strapping the socket on might be worse. Instead of adding mass, you want to be adding surface area, which is why sandwiching a thin plate between the sensor and o2 bung works well.

You're right about the air pocket, but there is plenty of turbulence in there to be clearing the air out. The socket is half open design. Plus the steel is a plenty fine heat conductor for this purpose. I'm quite familiar with the benefits of Al/Cu but I just don't see it being required if I can just strap my socket to the sensor. It's not a time sensitive thing, I don't much care if my heatsink takes an extra minute to reach nominal temp, etc. I'd also, likely have tripled the surface area along with the mass.

It's really just an idea that popped into my head. It's not glamorous or tested.
But I'd be happy to pit it against a thin piece of Al/Cu stuffed between the o2 housing and the sensor body.

EDIT: I may be thinking about this the wrong way. After reading it again, it seems that you guys are less concerned with the heat inside the 02 housing, but rather than heat radiated from the 02 housing. Is that correct? Cause in that case my "free" heatsink mod might not come out on top when pitted against a thin metal plate - of nearly any material. If that's what this is a about, then I'm betting even a soda can-heatshield would be better.

EDIT II: Just got Foxit running on this machine, so I couldn't view the pdf last night. I''m understanding much better now why the socket idea isn't such a hit. It IS about radiated heat from the housing. Well not exclusively, but as far as we care - yes. I don't see anyone messing with bung-extenders to get internal temps down. The heatSHIELD (sink doesn't seems to apply very well for me in this instance) ought to have a significant impact on total heat absorbed by the sensor.

 

[jcarmichael]

i ended up buying a 8 dollar round with an opening in the center piece of copper, cut it in half and cut the round center piece off. then used a 3/4th spade bit for my center hole for the actual sensor to go through. have a completely different problem now.

http://www.dsmtuners.com/forums/tuning-engine-management/428496-mtx-l-calibration-help.html

posted the link to my next question as I don't want to post the same question in two different threads.

if anyone can help, I searched on google and found some answer but nothing yet that has been helpful.

You're right about the air pocket, but there is plenty of turbulence in there to be clearing the air out. The socket is half open design. Plus the steel is a plenty fine heat conductor for this purpose. I'm quite familiar with the benefits of Al/Cu but I just don't see it being required if I can just strap my socket to the sensor. It's not a time sensitive thing, I don't much care if my heatsink takes an extra minute to reach nominal temp, etc. I'd also, likely have tripled the surface area along with the mass.

It's really just an idea that popped into my head. It's not glamorous or tested.
But I'd be happy to pit it against a thin piece of Al/Cu stuffed between the o2 housing and the sensor body.

EDIT: I may be thinking about this the wrong way. After reading it again, it seems that you guys are less concerned with the heat inside the 02 housing, but rather than heat radiated from the 02 housing. Is that correct? Cause in that case my "free" heatsink mod might not come out on top when pitted against a thin metal plate - of nearly any material. If that's what this is a about, then I'm betting even a soda can-heatshield would be better.

EDIT II: Just got Foxit running on this machine, so I couldn't view the pdf last night. I''m understanding much better now why the socket idea isn't such a hit. It IS about radiated heat from the housing. Well not exclusively, but as far as we care - yes. I don't see anyone messing with bung-extenders to get internal temps down. The heatSHIELD (sink doesn't seems to apply very well for me in this instance) ought to have a significant impact on total heat absorbed by the sensor.

yea the objective with the heat sink is to keep the o2 sensor from over heating the o2 housing itself, is going to reach high temps and its suggested you put the o2 sensor further down stream. however the closer you can get it to the motor exhaust wise the more accurate your reading is going to be. this is my first wideband so as far as the accuracy ranging I don't know if its a broad range or short range.