Reflow Profiling

Hi I have been conducting my first reflow profiles. Most of the texts that I have read suggest soldering the thermocouples to a populated pcb. I am finding this very difficult. I have been using a high melting point (Sn5 / Pb95) solder and the difficulty is in getting the thermocouple junction wetted. I have even tried using an active (acid) flux without success. The only way that I managed to make a joint was to end up with a large ball of solder at the point of contact between pad / leg / thermocouple. I have set my profile under these conditions but it does not seem correct. Any comments, suggestions on connecting thermocouples please

Regards

Terry

reply »

| Hi | I have been conducting my first reflow profiles. Most of the texts that I have read suggest soldering the thermocouples to a populated pcb. I am finding this very difficult. I have been using a high melting point (Sn5 / Pb95) solder and the difficulty is in getting the thermocouple junction wetted. I have even tried using an active (acid) flux without success. The only way that I managed to make a joint was to end up with a large ball of solder at the point of contact between pad / leg / thermocouple. I have set my profile under these conditions but it does not seem correct. Any comments, suggestions on connecting thermocouples please | | Regards | | Terry | Terry, I was having the exact same problem. I purchased Temprobe thermocouples from Sanders Technology in Hollis NH. These are a sealed thermocouple with a built in positioning/clamping device. You would clamp the device to the side of your PCB, position the probe into wet solder paste and lock it into place. You will need to use wet paste to get an accurate profile. This is the only means to which I have found to accomplish this. I ran tests between reflowed and non reflowed PCB's and did find differences in the profile. Hope this helps

Christopher Lampron

reply »

| Hi | I have been conducting my first reflow profiles. Most of the texts that I have read suggest soldering the thermocouples to a populated pcb. I am finding this very difficult. I have been using a high melting point (Sn5 / Pb95) solder and the difficulty is in getting the thermocouple junction wetted. I have even tried using an active (acid) flux without success. The only way that I managed to make a joint was to end up with a large ball of solder at the point of contact between pad / leg / thermocouple. I have set my profile under these conditions but it does not seem correct. Any comments, suggestions on connecting thermocouples please | | Regards | | Terry | Hi Terry. The three most popular methods for attaching T/C's to PCB's are: a. High Temp. Solder b. Kapton tape c. High temp. epoxy. You have chosen the most popular method and are experiencing the same difficulties as everyone else. So don't feel as if your all alone on this! For myself, the best technique I have is to use the smallest gauge T/C . I use 0.005 wire (X2) which yields a weld of about 0.013 to 0.015 inch. As you mentioned, solder alloy will not bond to Chromel or Alumel (I assume your using K type T/C's). Therefore, you must capture the wire within a solder joint. This can be difficult but the smaller the wire weld will determin how much solder is required to capture the wire. Oftentimes there is a via near by which you can insert your T/C and then "solder" it into the hole. This will get you into the game without your T/C wiggling out of a solder joint which will never wet with T/C alloys anyways (at which point you have to start over). Check out Omega they are the largest manufacturer of T/C's in the world. Good luck.

Phillip Hunter Engineering Support Technician Laughlin Wilt Group Inc

reply »

Hi Terry,

I do the same thing Chris is suggesting. I am not sure about the company who made the thermocouples I use but most probably it's the same. I'll let you know if they're different so that you'll have an alternative source.

I also agree with the populated board scenario for your profiling.

One last thing: make sure you change one thing at a time :-)

Regards

reply »

| Hi | I have been conducting my first reflow profiles. Most of the texts that I have read suggest soldering the thermocouples to a populated pcb. I am finding this very difficult. I have been using a high melting point (Sn5 / Pb95) solder and the difficulty is in getting the thermocouple junction wetted. I have even tried using an active (acid) flux without success. The only way that I managed to make a joint was to end up with a large ball of solder at the point of contact between pad / leg / thermocouple. I have set my profile under these conditions but it does not seem correct. Any comments, suggestions on connecting thermocouples please | | Regards | | Terry | Hi Terry,

Yes, there is a trick to geting these thermocouple beads into a solder joint. It's taken me a few years to figure it out, but I've finally gotten it down.

First, the smaller beads are easier than the larger ones, but I have a KIC and usually use the larger ones without much difficulty.

That high temp solder's a bear to work with, ain't it? The second you pull the iron off it freezes, and don't even try using conventional wick with it. So you only get one chance per joint to get it right. If you muff it up, just go to another joint.

Here's my Not-So-Secret-Anymore Method:

1) Pick a joint at the very end of the device. If you're right-handed, this is the last two leads on the right side of the plcc or soic. If you don't take these, you might just make one big high temp bridge up the entire side of the device. Use a large tip for your iron to make your life easier.

2) Get your thermocouple ready. No kinks or bends near the junction bead. Tin the iron with the high temp.

3) Hold the iron on the lead, and with your left hand, run the high temp solder in 'till you get a nice little blob. Pull the iron back SLIGHTLY (I mean ever so slightly), keeping a nice web of wet solder between the leads and the iron.

4) Now insert the thermocouple bead into the wet blob with your left hand.

5) CRITICAL STEP> Drag the iron and molten solder over the thermocouple to the left, encapsulating the bead.

Now, obviously if you are left-handed, reverse everything.

This is a foolproof way to attach thermocouples. I used to have a lot more difficulty with them until I figured it out. It would take me half an hour to put down 6 thermocouples; now it takes about 2 minutes. I teach engineer trainees with very little hand soldering experience and they pick it up very quickly.

I wish I could show you instead of describibg it, but after a few tries, you should be a pro.

reply »

| Hi | I have been conducting my first reflow profiles. Most of the texts that I have read suggest soldering the thermocouples to a populated pcb. I am finding this very difficult. I have been using a high melting point (Sn5 / Pb95) solder and the difficulty is in getting the thermocouple junction wetted. I have even tried using an active (acid) flux without success. The only way that I managed to make a joint was to end up with a large ball of solder at the point of contact between pad / leg / thermocouple. I have set my profile under these conditions but it does not seem correct. Any comments, suggestions on connecting thermocouples please | | Regards | | Terry |

Use of high temperature solders for attaching thermocouples can provide inaccurate temperature readings. In a thermocouple there will be two wires, of different metal alloys. One wire has a negative DC voltage and the other wire has a positive DC voltage. The resistance of the blended alloys at the thermocouple weld provides the reference for the thermal measurement. When solder is used to attach the thermocouple to the board, you risk shorting the two thermocouple wires through the solder. Solder is a better electrical conductor, than the alloys used in thermocouple wires. The reference reading will then be through the solder, rather than the thermocouple weld. Since your chart recorder, or thermal data recorder is calibrated for the alloys used in your thermocouple,(ex. type K), shorting through the solder will provide false readings. The applications people at Omega Engineering 1-800-826-6342 can give an excellent, over the phone tutorial on thermocouple attachment. The recommended method is to use a thermally conductive, electrically insulative epoxy. The one I prefer is sold by Dexter Corp. It's called Epoxi-Patch 9340 Gray. Phone number for Dexter is 603-474-5541. The 9340 Gray is a two part epoxy which cures at 125� C within minutes. It's easy to use, and will typically out last the profile board.

reply »