Best Re-flow Oven for Hobby electronics

[FotatoPotato]

Hi all, I'm starting work on a pcb that will have a 500 ball BGA chip and I need a re-flow oven in-order to solder it to the board . I did look on amazon but all the ovens I saw looked cheap, were from random Chinese sellers and the reviews were not that food so I was just wondering If you guys know of any good quality re-flow overs for under $500

Thanks 

[Simon]

I presume the cheap Chinese ovens are T962/T962A/T962B. Those are shit and those come with severe thermal design issues. They can't cook large boards evenly.

The cheapest good oven I would consider is T962C, the C version is completely redesigned with IR+hot air heating. It won't fit in your budget, but it's just a little stretch.

I have one of these at work, the controller is exactly the same as the smaller ones with 1s on the display actually being 1.5s, I dont see any hot air ? but it does have move even heat with more bulbs closer together. On mine one of the two fans has failed which is not a bad thing as the one left pushes heat around a bit and not much out.

I have another oven myself now made by a another company and new on the market, it does have an internal fan that moves air around as well as the exhaust fan.

[chris_leyson]

The toaster oven from Beta Layout is €389, or €489 with the magnetic stencil workbench. https://uk.beta-layout.com/estore/order_product_details.html?p=740
Mine arrived this week but I haven't tried it out yet, I will have to reflow some boards with QFN packages and didn't fancy risking a hot plate and hot air. I've done a few repairs with hot air in the past and it isn't easy. Dave reviewed the same oven a while back and gave it a thumbs up.
together with a follow up video


There is also an article on their website about soldering BGA packages in a pizza oven, in German.
https://uk.beta-layout.com/download/rk/RK-10590_702.pdf. It's important to use a heat spreader underneath the PCB so I'm going to try some thin brass sheet. Will let you know how it goes.

EDIT: Oven is 230V but it should work OK on 220V split phase.

[Simon]

I dont see any hot air ? but it does have move even heat with more bulbs closer together.

I think I read somewhere that it has internal fans as I was researching on getting myself a reflow oven. I didn't end up getting this one, though.

Well haven't taken one apart but I don't believe it. They just make 3 sizes of the same basic design.

[KE5FX]

I recently bought a T962A and added the ESTechnical aftermarket controller.  Not a whole lot of experience with it yet, but I've used it to solder a board with a 400-pin Searay connector and it seemed to do OK. 

The ESTechnical controller makes a huge difference.  Moving the graph to the PC is helpful in itself.  The T962 can't control the temperature of the board very closely, but the graph allows you to tweak the process to get something resembling the desired slope, e.g.:



The USB interface is flaky and requires frequent reconnection and/or power cycling, but the controller itself is solid.  It really makes the most of the oven's limited hardware design.  In a couple of weeks I'll run a board with an 0.8-mm BGA through it, and we'll see how it holds up.  But I have no reason to think it won't do OK.

It's important to use a heat spreader underneath the PCB so I'm going to try some thin brass sheet. Will let you know how it goes.

I haven't heard this before, any pointers?

The advice I've read (and followed with the Searay board) was actually to elevate the board in the drawer slightly by putting a couple of PCB scraps beneath it.  The ESTechnical controller uses the fan to provide air circulation during operation, so this seems like a good way to ensure the whole board is preheated adequately.

[jmelson]

I've been using a large GE toaster oven from Wal Mart for over 10 years.  It has a wire rack in the middle, and two horizontal elements below, and two above the rack.  While the heat distribution is not perfect, it seems to be within maybe 10 - 15 C worst case across all the boards.  i often do 4, 6 or even 8 smaller boards at one time, and have found a peak temp of 247 C is good for lead free, and 230 C is good for leaded solder.

I poke a micro thermocouple into a plated-through hole in one of the PCBs to read the temperature.  I got a ramp and soak temp controller on eBay, this allows you to program temperature ramps and hold at certain temps to get the desired profile.

I have done well over 1000 boards with this setup.

Jon

[flydrive]

I have another oven myself now made by another company and new on the market, it does have an internal fan that moves air around as well as the exhaust fan.

I know this is an old post but do you mind saying what make of oven you were talking about here nearly a year ago? One with an internal fan would be great, else I'm modding a T962A or even C.

[sairfan1]

I'm hobbyist, not very extensive user.  I found this oven way better than other available options. There are some good users in the thread i hope they will also give their input.

[Kasper]

I have worked at 3 places that use toaster ovens for soldering prototypes. They all used basic ovens (2 were IR, not sure about the third) and manually changed the temp every 1 minute.  Put the PCBs on a piece of paper on a metal tray. It seemed to work fine but I never did a thorough test of the solder quality.

If you just want to try it cheaply before you spend hundreds of dollars, buy a used toaster oven off your local classifieds. If you are near any post secondary school, there should be lots for sale.

[Simon]

Mine is made by Huaqi Zhengbag and "Zhengbag" is a good description of the noise my fuse board made when I plugged it in. After taking it apart I improved the earthing fearful for my life and after an argument with them they assured me they were taking my advice about earthing each metal panel and ensuring better clearances. Mine is the ZB2520HL

[Peabody]

I'm finishing up a project that uses the $10 Walmart hot plate and a $5 circular saw blade as a reflow device for smaller boards.  See attached.  While I did design a controller, it doesn't need one.  You just turn on the power twice during the relfow process for specific periods of time, and it seems to reflow fine with no overshoots, and nothing is scorched.  It requires a minor mod to the hot plate.  I hope to have a Github repo available, and possibly a video, within a week or so, and will post when that's done in case anyone is interested.  It is the lowest cost, simplest system I've seen.  I guess larger boards could possibly be done by moving the board around during the reflow process to "share" the center cool spot evenly.  And obviously, it doesn't work for boards with SMD components on both sides.

[flydrive]

I'm hobbyist, not very extensive user.  I found this oven way better than other available options. There are some good users in the thread i hope they will also give their input.

sorry but which oven do you mean, there's been a few mentioned in this thread so I couldn't work out which you have.

[flydrive]

Mine is made by Huaqi Zhengbag and "Zhengbag" is a good description of the noise my fuse board made when I plugged it in. After taking it apart I improved the earthing fearful for my life and after an argument with them they assured me they were taking my advice about earthing each metal panel and ensuring better clearances. Mine is the ZB2520HL

Thanks - I'd seen those but couldn't find one single review anywhere saying whether they were actually significantly different from the T962* series, especially with regards their advertised claim that they are 'convection + IR'. They look similar enough I've been assuming they too just have a 4-bar electric fire at the top and use the fan only for cooling. Clearly they suffer from some of the same build issues.

I already have the toaster oven setup and it's worked just fine for leaded work up to the size the thing fits. However I'm moving onto larger boards, 300x250 or so and some limited BGA work isn't far away. Toaster ovens I have tested which are large enough to take a board that size thus far either don't get nearly hot enough, or don't get there fast enough. They tend to triple the volume and add 50% to the power as far as I can tell. Hence my search for something custom designed for PCB work, even if it's going to require modification.

[wilfred]

I found this toaster oven for A$20 at Officeworks. I like it because it is small with the elements closer to the tray than is possible in larger ovens. It is 650Watts. Does anyone have experience with it for reflow? I'll take opinions too. It is super cheap but doesn't hurt to ask first.

Thanks

https://www.officeworks.com.au/shop/officeworks/p/heller-professional-stainless-steel-6l-toaster-oven-hpovt6#reviews

BTW I saw this one at Kew East store.

[FriedMule]

I think that the absolute best, non pro, home tinkerer oven, is one with forced air.
The ovens that are more like a toaster with glowing heating-elements is in my opinion, fare from ideal. Try to get a "real" oven, like a combo oven with forced air.

[jmelson]

I found this toaster oven for A$20 at Officeworks. I like it because it is small with the elements closer to the tray than is possible in larger ovens. It is 650Watts. Does anyone have experience with it for reflow? I'll take opinions too. It is super cheap but doesn't hurt to ask first.

Thanks

https://www.officeworks.com.au/shop/officeworks/p/heller-professional-stainless-steel-6l-toaster-oven-hpovt6#reviews

BTW I saw this one at Kew East store.

Well, this looks similar to one I have used for over ten years.  Mine has two heating elements that run left-right above the tray, and two below.  I spliced in a thermocouple ramp-and-soak controller in series with the thermostat.  I poke the thermocouple into a plated through hole in the board, so the control loop is actually controlling temperature on at least one spot of the board.  It has worked very well, and I've done close to 2000 boards with it, I think.  Mine has a convection fan in it, but it doesn't really move much air at all, so I don't think it actually accomplishes anything.

Jon

[Simon]

Air is a poor conductor of heat so moving the heated air to the board makes a lot of difference.

[wilfred]

I went and bought one. But I certainly don't recommend it for food. When starting from cold it goes to 295 Celcius and switches off after 4min22sec. Then it cools to 136 Celcius and switches on. It takes a minute and a half  to get back to 295 Celcius. And the cycle repeats.

I set up a DMM to measure the temperature and a video camera to watch it. The temp probe (EEVBlog 235 DMM) was in free air just above the wire rack.

I'd take it back if I wanted it for food. There is another review on the Officeworks website which said something similar so I don't think my unit is particularly defective. It's still only two samples of course so it is hard to read too much into that. But you'd think it would stay on for the entire 15minute timer cycle and vary the temp by a lot less than 160C.

But since I don't want it for food and intend to put a PID temperature controller to drive it I thought I'd seek further opinions. I'll see if I have time to plot a graph of the time/temp tomorrow. BUt for now I can say at full power it will heat from 136C to 295C in 94 seconds.  I'll try manually holding it from 250-270C and see if that doesn't trip the thermal cuttoff. Do you think that is satisfactory for solder reflow?

I think it might be OK but I've never built a PID controller or a reflow oven.

If it is OK I can say that since it is about the size of a loaf of sliced bread it would be good for a space constrained lab that does only one or two boards at a time. It seems pretty well built. It just doesn't work for its intended purpose.

Edit:
Here is a plot of time/temp.

[Simon]

I have a little fan oven that i use for food, i have noted that it is very "loose" on the temperature control as turning it up after it has cut power to the element on the current setting does not turn the element back on so it must be greatly overshooting the set temperature and have a big hysterysis. Works fine for food though.

[Doctorandus_P]

For me I would not consider a pre-built oven like the T962.
I've heard some bad reveviews about un-even heating, but those reviews are 10+ year old and with a new version those reviews are unreliable.

The main reason I would not buy such an oven is that you can not see what is happening with your solder job.
And if you can not see what is happening, you can not adjust to improve it and you can only see the end result.

For me I can see a few options.
I would consider a hot plate, which heats the PCB from the underside. The cheapest version of this would be a simle frying pan on a kitchen stove. I would not reccomend this for long term use, but I've heard success stories about it and it seems like a good idea to try it to get some experience on how it works. There are some relatively cheap hot plates directly from Ali / Ebay / China, and there are also some on Tindie, or started with a kickstarter.

Another option that looks good is to use a (kitchen) oven with a glass window. These need a fan for heat distribution, and you have to upgrade the temperature control with a thermocouple and a temperature controller, but there are lots of opern source projects of these (check github) and also kits and pre-build controllers.

[Simon]

My Zhenbang oven is a marked improvement. It has an air circulating fan. All of these IR ovens have a tiny glass window that has always worked for me. I just did some really heavy boards where I had to watch for it to reflow as it was taking longer than the standard profiles allows.

[Doctorandus_P]

Simon says exactly why you want to be able to monitor the soldering process, either via a "frying-pan" option or a glass window in the oven.
Different boards need different soldering profiles, and with a closed oven this is much more difficult to get right.

[jmelson]

I went and bought one. But I certainly don't recommend it for food. When starting from cold it goes to 295 Celcius and switches off after 4min22sec. Then it cools to 136 Celcius and switches on. It takes a minute and a half  to get back to 295 Celcius. And the cycle repeats.

Yes, the built-in thermostat is pretty awful, it is more of a power regulator than an actual temperature measuring device, at least on mine.  I never expected it to work, and immediately bypassed it with a thermocouple temperature controller and solid state relay.

Well, the problem is you want a controlled temperature of the BOARD, not the air in the oven.  I didn't know until I started doing this, there is almost NO RELATION between the two.  The air heats slowly (I guess due to the thermal mass of the oven's walls) but a board in the middle will heat quickly, due to IR transfer.  So, I got a big roll of an odd (Type E) thermocouple extension wire and a ramp-and-soak controller on eBay.  The micro-size thermocouple wire can be poked into a hole in the PCB to measure actual PCB temperature.  The ramp-and-soak controller allows you to program temperature points and time between them, so you set it like :
10 C : 3 minutes : 180 C : 1 minute : 247 C : 1 minute : 10 C : 2 minutes

Jon

[wilfred]

I added the graph of time/temperature to my earlier post.  It looks like the oven can heat up hot enough and fast enough to work.

The comment about measuring the temperature is one that has always intrigued me. How do you measure the temperature properly to do solder reflow. I'm certainly not claiming any expertise here. I've seen both probe in free air and inserted or in close contact with a PCB. Either way the temperature reading must be the actual probe tip temp. Both seem to work for melting the solder as does just sticking the PCB in a frypan and switching it on.

I suspect the profile is more relevant to the survival of the components. But then there are different profiles for different solder.

[jmelson]

I added the graph of time/temperature to my earlier post.  It looks like the oven can heat up hot enough and fast enough to work.

The comment about measuring the temperature is one that has always intrigued me. How do you measure the temperature properly to do solder reflow. I'm certainly not claiming any expertise here. I've seen both probe in free air and inserted or in close contact with a PCB. Either way the temperature reading must be the actual probe tip temp. Both seem to work for melting the solder as does just sticking the PCB in a frypan and switching it on.

My first test was with the thermocouple in air.  The boards burned while the temp indicated 230 C or so.  So, obviously, the boards absorbed more IR than the thermocouple.  Next test, I poked the thermocouple into a plated through-hole in the board, and got beautiful reflow soldering.  So, unless you attach the thermocouple to some kind of PC board simulator that has well-matched IR characteristics, you won't get the right temperatures.

Jon