hamster

Reflow oven

Recommended Posts

PCBs arrived today - spent the night doing some baking. It worked!

 

Eight_led_wing.jpg

 

(Well, except for the first board where I put all the LEDs on backwards... opps!).

 

The only problem is that some of the LEDs are not sitting flat - any hints? are the pads too small? too much paste?

 

The LEDs are larger than expected, and all the resistors are OK...

Share this post


Link to post
Share on other sites

It's probably too much paste causing the LEDs to not be flat. Are you using a stencil?

 

I usually use a dental pick to push down on a component while it is on a hot plate, that will usually push any excess solder out the side and make it sit flat.

Share this post


Link to post
Share on other sites

Hi I just incidentally run across Dave's latest EEVLOG vids regarding a simple reflow oven kit from germany :)

(Actually this are two links (one shown as a link one shown as a embedded youtube video )

 

http://www.youtube.com/watch?v=FNNRoXZom30

 

 

 

 

probably it's already seen but I thought posting this does not hurt

Share this post


Link to post
Share on other sites

Mike, I was wondering why you didn't line up the LED/resistors vertically and closer together (basically in line with the wing pins) like

 

Pin-- RES--LED

Pin-- RES--LED
Pin-- RES--LED

Pin-- RES--LED

etc

 

The board would have ended up being a fraction of the size not extending so much out (I have a thing for tiny minimalistic boards :) )

Share this post


Link to post
Share on other sites

The boards were the same price upto 50x50mm, so size wasn't an issue.

this allowed me to put them at the same spacing I am looking to use with some switches.

do you want me to drop a kit in the mail for you? might be useful for debugging... what am I going to so with 21 of them??

Share this post


Link to post
Share on other sites

Just to close this off too - did another couple of wings today, using the smallest amount of paste that I dared.

 

Got much better result. Everything flush and neat.

 

Third_run.jpg

Share this post


Link to post
Share on other sites

It is the least expensive oven from "The Warehouse" (perhaps NZ's equivalent of WalMarl) http://www.thewarehouse.co.nz/red/catalog/product/Necessities-Toaster-Oven-White-9L?SKU=1689799

 

Apparently one key thing is to make sure it has a full width tray in it - it helps defuse the heat. from the lower element, which can get shadowed by an oven rack giving cols spots.

 

I think that cheaper  is better - they have smaller doors (so give a more even heat than something with a lot of glass) and the lighter gauge of metal used most likely heats up quicker than a heavier oven would. 

Share this post


Link to post
Share on other sites

Ah crap, shopping time.... at that price I have to get one too :) The one thing they don't mention is the wattage, what does yours say?

Share this post


Link to post
Share on other sites

all I have seen so far is small stoves, with hotplates on top, I wonder if there is not allowed to sell such devices where the heating element is bare like that, or I'm looking for the wrong thing.. 

Share this post


Link to post
Share on other sites

1000W. of the sticker on the chassis. so a 40A SSR is overkill!

 

Chur Mike for posting some PCBs, must build the Oven setup first. Gotta shop for a cheap SSR and a K-type thermal probe + ADC and all the other bits and bobs first. Good project for next year (aka tomorrow, or depending on hangover, the day after that) especially since you did all the hard work figuring out everything and writing the HDL to drive it.

 

I was leaning towards this K type thermal probe, certainly the price is right. I like the threaded mount, means I can bolt it to something inside the oven or even through one of the wall plates.

 

Happy New year.

 

EDIT: OK shopping complete, Oven $30, MAX31855 based breakout board $25 (why so expensive?), K-type TC and 40A SSR for grand price of $8 so total $63 all up in local funny money shipping inclusive. All the other bits and bobs (FPGA, cables, etc) I already have. Not bad building a reflow oven for under a hundie.

Share this post


Link to post
Share on other sites

You might want to be careful with how you place the thermocouple.

 

Unlike when baking a cake, the oven isn't in a 'steady state' condition 

 

Back in High school IIRC heat travels by three methods - convection, conduction and radiation. To accurately gauge the temperature of the PCB and parts it needs to be:

 

- In the same 'air flow' as the  PCB - so convection is the same, not higher or lower in the oven, or at the edge of the oven.

- located on something of a similar heat to what the PCB is  - so the heat transfer by conduction is the same 

- Responds the same to IR radiation from the element - an OSHPARK purple PCB will heat quicker than a trendy 'all white' one.

 

I think a glass braided thermocouple like http://nicegear.co.nz/sensors/thermocouple-ktype-glass-braid-insulated/ that you can mount on a scrap PCB of the same type (using kapton tape, but not over the tip ) will get you better / more accurate results, even if it is a pain to use. 

Share this post


Link to post
Share on other sites

you've seen the videos? It's as hamster stated above. The reflow temperature is really critical in respect to the performance (or even survival) of components... be sure to stay away from the max. temps in the datasheets. So measuring temperature at the body of the oven may mislead and cause real trouble

Share this post


Link to post
Share on other sites

I've ordered a batch of PCB from Seeed (ETA 3 weeks). If anybody would like one (less the through-hole components) just drop us a line. The pins are laid out for the Adafruit Themocouple Amp/ADC board, but should work with andything. It also has four general purpose LEDs on it

 

Reflow_wing.png

  • Like 1

Share this post


Link to post
Share on other sites

Watch it with those cheap Chinese SSRs. Some of them are halfway decent, but often the ratings are vastly overstated. I saw inside one "30A" SSR that had a little TO-220 triac screwed to the aluminum backplate without any heatsink compound. Even the cheap ones ought to be fine for 1kW though.

 

I use a hotplate I made out of a piece of 3/8" aluminum with cartridge heaters inserted in holes I bored down the edge. It's mounted in a tray on a project box with a cheap PID controller and SSR in it and works well. The downside is that it only works to reflow parts on one side of a board but it's more compact and easier to control than an oven since it transfers heat via conduction.

Share this post


Link to post
Share on other sites

Yes my thoughts exactly. I pretty much always follow Dave's motto from eevblog, "don't turn it on, take it apart" and will try and check the ratings of the internal components. Of course sometimes the components can be re-labeled to show higher ratings...

Share this post


Link to post
Share on other sites

I've been known to xray such things on occasion. There is also the option of testing one out at the rated power and see if it blows up. An additional difficulty with the cheap knockoffs is that two identical looking assemblies may have very different internals.

 

Occasionally you can find a deal on some good name-brand SSRs in the sea of random stuff on ebay. You can also build one yourself easily out of a triac, optocoupler and a couple of resistors. When it comes to heating devices I don't like to take chances as solid state switches typically fail shorted = heater stuck on.

Share this post


Link to post
Share on other sites

Well despite all my intentions to disassemble the SSR-40DA Solid State Relay module, it had other plans. The little bugger is entirely potted in some hard black epoxy so the most I could achieve is to pop open the little plastic cover and peek inside helplessly. As I only bought one of these (heck at $5 I should have bought a few) a destructive teardown is out of the question at this stage.

 

post-29560-0-16674700-1389764856_thumb.j

 

To be honest given the price I was half expecting to see one of those

 

s202.jpg

just bolted to the rear metal plate. These are only rated to about 8A, so maybe the Fotek really can handle the stated current.

 

The other question I have which was not answered by the datasheet I found is whether the rear metal plate is isolated or not. With my DMM on 20Mohm range I could not find any continuity from the rear plate to any of the four pins in either direction (ie negative probe on the plate, then same test but positive probe on the plate). Might need to test again with mains going through it to be sure.

Share this post


Link to post
Share on other sites

I've never seen one that wasn't isolated. It's good practice anyway to ground the heatsink though rather than rely on the isolation unless the whole thing is fully enclosed. That way even if a wire comes loose or some other failure occurs, the heatsink can't become live.

 

If the thing runs for a few hours without failing or getting excessively hot, it will probably run fine for years. In the case of the reflow oven I wouldn't worry too much, it's not the sort of thing that will be running unattended for long periods. If I were trying to control a built in heating device that operates autonomously I would shell out for a name brand SSR and/or include a secondary failsafe and a thermal fuse as a last ditch protection.

Share this post


Link to post
Share on other sites

The Sharp four terminal SSR iin the second picture above has mains level voltages on its metal heatsink rear pad according to a blog where someone has used it in a reflow oven setup. This is why I was worried the other one might too.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now