DIY SMT Solder Oven

[Jacques St-Pierre]

Hello Group,



This may not be the appropriate news group to discuss this, but the group
helps me before and I do not know a better place to find help again.



As many, I am trying to build a low cost SMT Reflow Oven using an of the
shelves small oven. I did have one first prototype working, but it was too
small and plague with many problems, mostly cause by uneven heat diffusion.



I am working on s second unit, build with a lot more powerful heaters,
cooling fan, and more evenly distribute heater surface. With this one, I
have more than enough power to raise temperature to target value with the
proper raise speed.



Not to say, that the unit is control by computer, PID is use to achieved
proper temperature control. The PCB temperature is raise to 150c at a rate
of 2c/second, the PCB soak at 150c for 60 seconds and it is raise to 215c at
the same rate. Next temperature is cool down by a fan. Time over melting
point (183c) is around 60 seconds has require by the paste we used.



The problem is that with those timing, we end up with a toasted PCB.
Reducing the time over melting point, did correct the problem, but barely;
the PCB still have trace of over heating. Trying the same process with RoHS
paste, using a higher melting point temperature (220c), resulted in PCB burn
up.



I tried to implement convection, but the blower is not powerful enough, it
has no effect on the result.



I use a thermocouple in contact with the PCB to pick up temperature. I am
not sure if I should monitor the air temperature at proximity of the PCB or
still use direct contact? Both methods results in quite different
temperature response.



Do you have any hints to the reason why the PCB toast? What is wrong with
the method? Do I heat too quickly or too high?



I do over shoot a bit at the top, I should raise up to 215c, but end up at
225c for a few second before the fan turn ON, is this why the PCB have trace
of over cooked? In the case of the RoHS paste, I must rise to 245c; I
overshoot to 250c, but the PCB start burning well before I reach that
temperature.



I use standard fiberglass PCB. I must use and oven since I have SMT on both
side. I did solder many PCB with the first prototype, but they must be
small. The second unit will be able to older PCB up to 12" x 12", but I must
find a way to prevent burn on the PCB.



Need help and suggestions.



Thank you all.



Jacques

 

[[email protected]]

Hello Group,

This may not be the appropriate news group to discuss this, but the group
helps me before and I do not know a better place to find help again.

As many, I am trying to build a low cost SMT Reflow Oven using an of the
shelves small oven. I did have one first prototype working, but it was too
small and plague with many problems, mostly cause by uneven heat diffusion.

I am working on s second unit, build with a lot more powerful heaters,
cooling fan, and more evenly distribute heater surface. With this one, I
have more than enough power to raise temperature to target value with the
proper raise speed.

Not to say, that the unit is control by computer, PID is use to achieved
proper temperature control. The PCB temperature is raise to 150c at a rate
of 2c/second, the PCB soak at 150c for 60 seconds and it is raise to 215cat
the same rate. Next temperature is cool down by a fan. Time over melting
point (183c) is around 60 seconds has require by the paste we used.

The problem is that with those timing, we end up with a toasted PCB.
Reducing the time over melting point, did correct the problem, but barely;
the PCB still have trace of over heating. Trying the same process with RoHS
paste, using a higher melting point temperature (220c), resulted in PCB burn
up.

I tried to implement convection, but the blower is not powerful enough, it
has no effect on the result.

I use a thermocouple in contact with the PCB to pick up temperature. I am
not sure if I should monitor the air temperature at proximity of the PCB or
still use direct contact? Both methods results in quite different
temperature response.

Do you have any hints to the reason why the PCB toast? What is wrong with
the method? Do I heat too quickly or too high?

I do over shoot a bit at the top, I should raise up to 215c, but end up at
225c for a few second before the fan turn ON, is this why the PCB have trace
of over cooked? In the case of the RoHS paste, I must rise to 245c; I
overshoot to 250c, but the PCB start burning well before I reach that
temperature.

I use standard fiberglass PCB. I must use and oven since I have SMT on both
side. I did solder many PCB with the first prototype, but they must be
small. The second unit will be able to older PCB up to 12" x 12", but I must
find a way to prevent burn on the PCB.

Need help and suggestions.

Thank you all.

Jacques

I'd certainly like to hear about your project, so please post your
progress.

 

[[email protected]]

Put the TC on the metal of the heating element, if its on the PCB, you
are experiencing overshoot because of the PCBs low thermal
conductivity, perhaps?

Steve Roberts

 

[James Arthur]

Here's my thinking: Most of these cheap toaster ovens heat unevenly. Its
possible that your sensor is reading in a cooler spot than where your PC
board sits.

Also, make sure that the PC board is not exposed directly to the heating
elements or direct radiation will heat it too fast.

One idea I've read about is to incorporate a small fan inside the oven
to keep the air mixed and at a more uniform temperature. A small
impeller on a long shaft with the motor sitting outside the oven will
do.

This is the classic toaster-oven soldering guide:
http://www.seattlerobotics.org/encoder/200006/oven_art.htm


The Sparkfun soldering ideas (here's a hotplate article, with
links to the toaster project too):
http://www.sparkfun.com/commerce/tutorial_info.php?tutorials_id=59


Circuit Cellar article:
http://www.circuitcellar.com/library/print/0704/Lacoste_168/index.htm


Another, from Cornell, with controller:

http://instruct1.cit.cornell.edu/courses/ee476/FinalProjects/s2006/ki38/Webpage/index.html


The articles above discuss the various problems, solutions, and results.

HTH,
James Arthur

 

[qrk]

Hello Group,

This may not be the appropriate news group to discuss this, but the group
helps me before and I do not know a better place to find help again.

As many, I am trying to build a low cost SMT Reflow Oven using an of the
shelves small oven. I did have one first prototype working, but it was too
small and plague with many problems, mostly cause by uneven heat diffusion.

I am working on s second unit, build with a lot more powerful heaters,
cooling fan, and more evenly distribute heater surface. With this one, I
have more than enough power to raise temperature to target value with the
proper raise speed.

Not to say, that the unit is control by computer, PID is use to achieved
proper temperature control. The PCB temperature is raise to 150c at a rate
of 2c/second, the PCB soak at 150c for 60 seconds and it is raise to 215c at
the same rate. Next temperature is cool down by a fan. Time over melting
point (183c) is around 60 seconds has require by the paste we used.

The problem is that with those timing, we end up with a toasted PCB.
Reducing the time over melting point, did correct the problem, but barely;
the PCB still have trace of over heating. Trying the same process with RoHS
paste, using a higher melting point temperature (220c), resulted in PCB burn
up.

I tried to implement convection, but the blower is not powerful enough, it
has no effect on the result.

I use a thermocouple in contact with the PCB to pick up temperature. I am
not sure if I should monitor the air temperature at proximity of the PCB or
still use direct contact? Both methods results in quite different
temperature response.

Do you have any hints to the reason why the PCB toast? What is wrong with
the method? Do I heat too quickly or too high?

I do over shoot a bit at the top, I should raise up to 215c, but end up at
225c for a few second before the fan turn ON, is this why the PCB have trace
of over cooked? In the case of the RoHS paste, I must rise to 245c; I
overshoot to 250c, but the PCB start burning well before I reach that
temperature.

I use standard fiberglass PCB. I must use and oven since I have SMT on both
side. I did solder many PCB with the first prototype, but they must be
small. The second unit will be able to older PCB up to 12" x 12", but I must
find a way to prevent burn on the PCB.

Need help and suggestions.

Thank you all.

Jacques

I use a waffle iron with dual sided grill plates. One side is for
waffles and the other side is smooth for doing grilled cheese
sandwiches (I'm guessing).

I use the smooth sides of the plates. I put spacers in, if needed, so
the plates don't hit the components. This provides moderately even
heating from below and above. Put a dab of solder paste in a visible
area so you can see when the solder melts. Works fine for soldering
those pesky ICs with exposed paddle.

 

[James Arthur]

Put the TC on the metal of the heating element, if its on the PCB, you
are experiencing overshoot because of the PCBs low thermal
conductivity, perhaps?

Steve Roberts

This guy recommends aluminum heat shields to spread the
heat, avoiding hot spots and scorching:

http://ezinearticles.com/?Tools-and-Items-Needed-For-Soldering-Using-a-Toaster-Oven&id=921003

"You will need to make the following modification to your toaster oven
so that your boards don't get toasted by direct infra red radiation from
the heating elements.

The objective is for the heater elements to simply heat the air inside
the oven, so a direct line of infra red heat straight from the elements
to the circuit board must be avoided.

The way to achieve this is to get some thin sheet aluminium and make a
cut-out shape that just covers up the elements. You can use a disposable
baking tray from the supermarket to make this aluminium shield."


Cheers,
James Arthur

 

[JeffM]

[verbiage trimmed]
http://www.seattlerobotics.org/encoder/200006/oven_art.htm
http://www.sparkfun.com/commerce/tutorial_info.php?tutorials_id=59
http://www.circuitcellar.com/library/print/0704/Lacoste_168/index.htm
http://instruct1.cit.cornell.edu/courses/ee476/FinalProjects/s2006/ki38/Webpage/index.html

The articles above discuss the various problems, solutions, and results.

...and a couple more:
http://www.ustr.net/smt/index.htm
http://www.beloev.net/gbvio.html

 

[dalai lamah]

Un bel giorno Jacques St-Pierre digitò:

Do you have any hints to the reason why the PCB toast? What is wrong with
the method? Do I heat too quickly or too high?

How do you manage the cooling phase? With homebuilt reflow ovens (made with
toaster ovens, designed to keep the heat inside as much as they can) it is
normally necessary to manually open the oven door at a certain point, in
order to quickly dissipate the heat.

 

[Martin Riddle]

| [email protected] wrote:
| > Put the TC on the metal of the heating element, if its on the PCB,
you
| > are experiencing overshoot because of the PCBs low thermal
| > conductivity, perhaps?
| >
| > Steve Roberts
|
| This guy recommends aluminum heat shields to spread the
| heat, avoiding hot spots and scorching:
|
|
http://ezinearticles.com/?Tools-and-Items-Needed-For-Soldering-Using-a-Toaster-Oven&id=921003
|
| "You will need to make the following modification to your toaster oven
| so that your boards don't get toasted by direct infra red radiation
from
| the heating elements.
|
| The objective is for the heater elements to simply heat the air inside
| the oven, so a direct line of infra red heat straight from the
elements
| to the circuit board must be avoided.
|
| The way to achieve this is to get some thin sheet aluminium and make a
| cut-out shape that just covers up the elements. You can use a
disposable
| baking tray from the supermarket to make this aluminium shield."
|
|
| Cheers,
| James Arthur

This was my first thought, heat shields or diffusers.

Cheers

 

[Jacques St-Pierre]

How do you manage the cooling phase? With homebuilt reflow ovens (made
with
toaster ovens, designed to keep the heat inside as much as they can) it is
normally necessary to manually open the oven door at a certain point, in
order to quickly dissipate the heat.

At this point, I did add a fan connect to back of the oven with 2 feet's of
3 inches metal pipe. The fan is located lower than the oven, so this prevent
the heat from getting out by that opening. We also remove the bottom tray
(normally use to remove crumb), this give an large opening to exit the heat
when the fan turn on. We do not need to open the oven until the end of the
process.

Bye
Jacques

 

[Jacques St-Pierre]

"You will need to make the following modification to your toaster oven
so that your boards don't get toasted by direct infra red radiation from
the heating elements.

The objective is for the heater elements to simply heat the air inside the
oven, so a direct line of infra red heat straight from the elements to the
circuit board must be avoided.

The way to achieve this is to get some thin sheet aluminium and make a
cut-out shape that just covers up the elements. You can use a disposable
baking tray from the supermarket to make this aluminium shield."

At this point, all my reading appear to agree with this. In my last test, I
put a metal plate over the lower heater, this help. More modification will
be require to install shielding on both heater.

I also put the heat sensor in the air near the PCB, instead of direct
contact. This give more constant result.

I will try to install a fans inside the oven to move air around, this may
help a lot. My first try was not successful, the fan I use was not moving
air at all. I will try something else.

Bye
Jacques

 

[JosephKK]

This guy recommends aluminum heat shields to spread the
heat, avoiding hot spots and scorching:

http://ezinearticles.com/?Tools-and-Items-Needed-For-Soldering-Using-a-Toaster-Oven&id=921003

"You will need to make the following modification to your toaster oven
so that your boards don't get toasted by direct infra red radiation from
the heating elements.

The objective is for the heater elements to simply heat the air inside
the oven, so a direct line of infra red heat straight from the elements
to the circuit board must be avoided.

The way to achieve this is to get some thin sheet aluminium and make a
cut-out shape that just covers up the elements. You can use a disposable
baking tray from the supermarket to make this aluminium shield."


Cheers,
James Arthur

Hello, 12 inches by 12 inches is rather bigger than any toaster oven
that i know of. This guy has custom built his own. Adjust your
advice accordingly.

That said, i do not get why the small overshoot from 245 C for RoHS
solder to 250 C is damaging normal FR4. RTOP.

 

[James Arthur]

Hello, 12 inches by 12 inches is rather bigger than any toaster oven
that i know of. This guy has custom built his own. Adjust your
advice accordingly.

That said, i do not get why the small overshoot from 245 C for RoHS
solder to 250 C is damaging normal FR4. RTOP.

The man said his first unit had trouble with uneven heating. He
further relates that the 2nd unit burns boards before it even
gets to soldering temperature.

That also sounds like uneven heating and / or a direct IR path
that burns the board before the more-reflective solder gets up to
temperature.

ISTM 12x12 inches makes the likelihood of such problems even
greater.

So the cure (if that's the culprit) is a heat shield that blocks
the direct IR path.

Cheers,
James Arthur