Temp profile of reflow hot plate

Looks ok.. Google this:- reflow profile for leaded solder

Details are unknown, so it's hard to comment. Here's some things I learned from trying to reflow laptop boards...

The solder joint doesn't care one bit about the temperature of your saw blade.

All that maters is the temperature AT THE SOLDER. That can vary widely across the board depending on The thermal mass of the components. The radiation and convection off the top of the board in response to air currents. How well is the board coupled to the saw blade Whether you have ground planes or large conductor areas The list goes on...

I was trying to use hot air on the backside for preheat and hot air on the topside for soldering. I put some solder on the top of the chip so I could get some idea about whether the temperature was hot enough to melt solder.

You don't want to be moving the saw blade until the solder has solidified. I never found the chips that fell off the board when I jiggled it too much.

If I were to try again, I'd use a thermal imager. But that brings its own problems as the emissivity of various components introduces temperature errors.

Bottom line is that the shape of your curve is only part of the problem. The devil is in the details. If you've got a few 1206 or SOT-23 parts, it doesn't matter much what you do. When the solder turns from paste to silver liquid, you're done. Beyond that, it gets complicated rapidly.

A fan could speed up the cooling. It might also be used to help with overshoot, thus enabling quicker heatup. Duct the air via cardboard tube and you could maybe use a PC fan.

NT

Thanks very much for your reply. But first, this isn't intended to be used for laptop motherboards. It's for small hobbyist boards such as you might get for a few dollars from OSHPark for a DIY project. Small stuff.

In developing the profile, I put the thermocouple tip at a plated through hole on a small piece of PCB, hoping that would produce real-world results. And I hoped that the thermal mass of the saw blade and heating elements would overshadow any variation in the mass of boards or parts. Like cooking with cast iron. But I could add a push button and LED indicator to let you choose one of several slightly hotter profiles than my default - for bigger boards or colder ambient temperature, or I guess even for non-leaded paste.

Since the heating elements aren't quite all in a plane, I replaced three screws with longer ones, so that the blade is actually about 3/8 inch above the elements. That has the additional advantage of helping to even out the temperature variations between blade sections. I've tested my profile directly above an element, and midway between two elements, and the results are very close. That was encouraging.

I don't have any paste, but put regular 60/40 solder on a piece of pcb, and it melted ok fairly early on in the reflow stage. Of course that's not the same as paste, but a good indication.

What I see online in the Youtube videos on hot plates, amazingly enough, is guys removing the boards with tweezers immediately after reflow while the solder is still molten. I figured if they can get away with that, moving the saw blade should work too. But I certainly understand your point.

I deliberately chose the thermocouple because I thought it would give me more accurate readings than an IR temp reader

- because of the emissivity problem and the issue of not really knowing how wide the aperture is, I used a TM-902C, which is pretty cheap, but seems to work quite well,

To be honest, I was really surprised that people used hot plates at all for reflow. It just seems all wrong. But there is a large group of experienced people who strongly prefer hot plates to toaster ovens. They say the ovens have hot spots and cold spots, and they often produce poor results - scorched boards or parts, un-flowed sections, and melted connectors. And you can't really see whats going on. Also, the placement of the thermocouple is crucial with an oven, and has to be done right every time. You can't just measure free air in the vicinity of the board. With a hot plate, they usually don't measure at all, just turn it on, then remove the board after the reflow. I was trying to be a bit more elegant than that and at least follow the Kester profile more or less.

Well, I'll have to do some tests with leaded paste and sample boards and parts, and see how well it does. Maybe there are kits for that on Ebay.

There are a number of commercial "DIY" controllers for toaster ovens, complete with PID controllers, displays and other fancy stuff, and the world doesn't need another one from me. But you're going to be over $100 for a project like that, more likely over $200, and a major investment in time. I just thought it might be possible to do something really easy, quick and cheap that would work well enough - if I made the investment in time to develop the sequence. Others would only need an Arduino, a hot plate, a saw blade, an SSR, and some kind of housing. Oh, and the firmware which I would post on Github. They could invest in a TM-902C if they like, or an IR thermometer, but that shouldn't be necessary. And of course with the firmware sketch, they could always develop their own duty cycle sequence if they don't like mine.

I think we're talking about different devices

The problem is that you want to control temperature, but all you have is energy.

Radiation gives you great thermal transfer, but it's extremely dependent on the emissivity of the source/target surfaces.

Convection is vulnerable to local air currents.

Conduction can give you very precise control for things that are in intimate contact, but it's very sensitive to small changes in that contact area.

What you want is a source of low temperature and high conductivity to the target.

I know nothing about current technology, but 30 years ago, vapor-phase reflow was all the rage. You hit it with a vapor that condensed at the temperature you wanted. Huge heat transfer, correct temperature, self compensation for thermal mass. And contamination! Parts were a lot bigger back then.

What I tried to do was hit the backside with a lot of airflow at the soak temperature, then hit the front side with hot air at the flow temperature.

I found that a heat gun with a variable heat output and manual control using a thermometer was adequate.