How to rework small surface mount LED

I talked about soldering small surface mount LED in my previous posts by using DIY reflow oven. Today I would like to show you how to rework similar LED, which is, to desolder this LED off the PCB and solder a new one on it.

It is quite different from populating a bare circuit board because the board comes with all types of other components. Rework process is more to fix a problem than to create new damage, so I will need to focus on the target LED only, not to affect other ICs. The LED is from Philips Luxeon Rebel series, about 3mm x 5mm in size. The pads for soldering are on the bottom of the component, making it impossible to solder with the traditional soldering iron

Heat gun is a great tool for this purpose as it is good at spot heating. With proper procedure and handling, the removed LED might still be reusable.

Stencil, unfortunately, couldn’t be used in this situation to assist the solder paste applying process. However, thanks to the surface tension, manually applying the paste onto the solder pads still work for this job. Don’t worry if they look messy at the beginning.

Check out my video for more details!

The one with yellow bulb was just reworked.

Teardown of Ricoh Theta 360 Degree Spherical Panorama Camera

The first impression of the Ricoh Theta 360 Degree Spherical Panorama Camera is its odd appearance — to me it feels like a bar of chocolate wafer. The dual lens reminds me of the bubble eye goldfish. For some reason I was given this 1st generation Ricoh Theta camera for free, but broken. So why not taking a peak at its inside?

The camera was wrapped with aluminum straps by double-sided tapes to cover the screws. Removing the straps would expose the screws. And that would be the starting point for the teardown.

Siding removed.
Siding removed.

There were quite a few screws on the sides. Once they were all gone, the cover popped up easily. Now you can see what’s enclosed: battery; circuitry and optical module. Looks like each section takes equal amount of space inside the camera body.

Open up the body

The Li-ion battery (model: DB-100) is quite beefy, to be honest. It is actually a Ricoh’s standard rechargeable battery product, which is also being used on CX5, CX4 and CX2 cameras. No wonder it looks a little over-sized for the slim panorama camera. I guess Ricoh could cut down the cost by doing so.

A big battery.
A big battery.

Panorama camera is equipped with a RS-WC-201 wireless module from Redpine Signals. This module supports 802.11b/g and single stream 802.11n, up to 65Mb/s. Retail priced at about $65.00USD from Mouser. Same feeling for the battery that it is also quite big….but off-the-shelf means lower cost, why not?

Wireless module RS-WC-201
Wireless module RS-WC-201

Unplugging the wireless module will expose the mother board and the optical assembly. The optical assembly, featuring two big fish-eye like lens in a back-to-back configuration, is well sealed and weighs a lot. A metal strap was wrapping around the lens, and was soldered on to the mother board through black jump wires. I guess that was some sort of anti-static protection since the camera design made it very likely for a user to touch the lens. Although the lens is a passive component, prone to electrical discharge, the enclosed image sensor (I will show you later) may get affected if the lens takes the charge.

Optical module.
Optical module.

The mother board is actually bigger than my first glance. It goes under the battery, covering the whole body except for the lens area.

Battery contacts on the mother board. The battery is actually riding on top of the mother board.
Battery contacts on the mother board. The battery is actually riding on top of the mother board.

Another interesting discovery from the back of the mother board is that the Theta camera is utilizing a 4G micro SD card as extra storage.

FPGA and 4G micro SD card on the back of mother board.
Spartan-6 FPGA and 4G micro SD card on the back of mother board.

A piece of metal shield covers the ARM processor and the memory chip. The ARM processor is Fujitsu Milbeaut Image Processor MB91696AM, equipped with a dual-ARM core and built-in H.264 codec that supports Full HD video. This processor seems to be 4~5 years old already, its maximum of 5.5fps at 20MP doesn’t look too fancy on today’s market. That’s probably part of the reasons why Ricoh has released a new generation: Ricoh Theta S. The memory chip from CHIPSIP model CT49248DD486C1 is NAND(1Gb)+DDR3(2Gb). So my question is, if the storage size could be easily increased by using a bigger NAND chip, what is the necessity for an add-on micro SD card? Maybe it is easy to make models for different memory size?

Metal shield covering the ARM processor and memory chip.
Metal shield covering the ARM processor and memory chip.

The Xilinx Spartan-6 FPGA, the dual-core ARM processor and the memory chipset make the image processing engine for the Ricoh Theta camera. What makes it different from all other cameras is the unique optical/image sensor assembly that is capable of capturing 360° panorama photo.

Side view of the optical / image sensor module in a compact package.
Side view of the optical / image sensor assembly in a compact package.
Top view of optical / image sensor module in a compact package.
Top view of optical / image sensor assembly in a compact package.

I took the photo below as I though this would be the end of my teardown work of the Ricoh panorama camera. But I was wrong! Please read on as more interesting discoveries are ahead about the well engineered optical / image sensor assembly.

Disassembled Ricoh Theta Panorama Camera.
Disassembled Ricoh Theta Panorama Camera.
Disassembled Ricoh Theta Panorama Camera.
Disassembled Ricoh Theta Panorama Camera.

The two circuit boards on the top and bottom of the optical assembly are actually the image sensor boards. I didn’t realize this until I further tear down this assembly. It was hard to image how a sensor is not directly behind a lens, but on the side looking through a diagonal mirror. What a smart design! Since the image sensor is no long taking up the space behind the lens, the back-to-back dual fish eye lenses could get as close to each other as possible. This, presumably, will leave a smallest as possible seam between the two 180° panorama photos. I guess this would make software engineer a much easier job to stitch the photos.

The optical / image sensor module was so well packed that it took me and my colleague some serious efforts to open it up. I noticed that some glue was used for the packaging, so the teardown of this assembly is an irreversible process. It was causing permanent damage to the assembly. So think twice before you try this.

The image sensor got knocked off first. I wasn’t able to tell its maker or the model number. Given the limited number of pins on the connector, I guess it is a rolling shutter sensor.

Top and bottom view of the image sensor.
Top and bottom view of the image sensor.

Once the sensor was removed, I was able to prove my assumption of its right angle optical path. As you can see from the photo below, I shined some yellow light on the right side, and the color showed up on the top. Same thing happened on the left side where the image was reflected down. There must be some sort of double-sided mirror placed in diagonal direction.

Image on the right gets reflected up, the left down.
Images from sides of the Theta panorama camera lens are reflected up and down to respective image sensor.

After the out-most convex lens was gone, we were able to take a further step into the heart of the optical module. But we were stopped by another piece of lens.

A second lens along the optical path
A second lens along the optical path

After another desperate digging, knocking and poking, the 2nd lens was finally gone. And the core of the optical module was just a friction of an inch away.

Lens module stripped to the core.
Lens module stripped to the core.

Finally we were able to grab the core: a beam splitter like cube that is made of two prisms glued together, forming a double-sided mirror on its interfacing surface (45°).

Theta Panorama Camera mirror cube.
Theta Panorama Camera mirror cube.
Theta Panorama Camera mirror cube.
Theta Panorama Camera mirror cube.
Theta Panorama Camera mirror cube.
Theta Panorama Camera mirror cube.
Lens design diagram of Theta camera

Here is the photo of all the parts that were ripped off from the optical / image sensor assembly.

Ricoh Theta Panorama Camera optical / image sensor disassembled parts。
Ricoh Theta Panorama Camera optical / image sensor disassembled parts。
Ricoh Theta Panorama Camera optical / image sensor disassembled parts。
Ricoh Theta Panorama Camera optical / image sensor disassembled parts。

Soldering a high-density LED array using DIY reflow oven

First of all, please watch the video below to find out how everything was done on my work bench.

A recent project requires soldering 600+ tiny LEDs on long narrow circuit boards. The LED, Philips Rebel Z ES, is as small as 1.6mm x 2mm, and is placed 2.5mm apart from each other. The leads of those LEDs are only on the bottom of the component, so apparently hand soldering is not possible.

Philips Rebel Z ES LED size vs. US quarter dollar
Philips Rebel Z ES LED size vs. US quarter dollar

Since I’ve built a DIY reflow oven before, I decided to push the boundary a littler further and give it a try. I’ve done many project using this DIY oven, but components’ density at this level is something I’ve never tried before. Carefully starting this project by designing proper circuit, I also tuned and laser cut the stencil, and finally got into this reflow process. Surprisingly, all those pieces worked together very well and got this job done!

How to rework high density package ICs with DIY reflow oven

I am working on a DIY PCB project that requires mounting three PICOR PI3301 Power regulators. Unfortunately, the PI3301 comes with a high density 123-pin LGA footprint in a 10mm x 14mm package. Furthermore, three of them will need to work in a synchronized close-loop configuration to yield high current output. That is to say, the power module won’t work efficiently unless all of the three DC-DC buck converters are properly soldered. How can I accomplish this with my re-flow oven?


My re-flow oven has been part of my workbench tool team for more than a year, and I had successfully re-flowed a variety of electrical components, as small as 2mm-wide the Philips Rebel Z LED. However, re-flow of a dense package IC such as LGA (or BGA) is something I have never tried before. So I carefully prepared  the stencil, applied the paste, placed the components, and Baked! Here is what I got from my oven:

Three PI3302 in synchronized parallel mode

Everything looked perfectly fine from outside. However, you just couldn’t tell if each 123 pads are correctly soldered without an X-ray. So I loaded the power module with dummy resistors to stress test the power supplying. After about 5 minutes, here was what I could tell:

The right one is not working.

The left two were getting warm; the one on the right stayed cool. Apparently the cool one is not functioning, most likely caused by bad re-flowing. So how can I fix it without affecting the left two that seemed to be working correctly? Or, how can I partially re-heat the right side of the board and re-flow again? Here is what I came up with:

Load part of the PCB into the oven.
Shake off the components when the solder is melting.

Apparent this is not a cool way of using an oven, but it works! Only the right side of the circuit board is inside the oven. Some PCB scrap and metal sheet were used in replacement of a “door” to cover the opening. Surprisingly, the oven heated up the board really well.

All I needed to do was to wait until my PID thermal controller heated up the board till 230°C, and then SHAKE it! Since the solder on the right side of the board was being melt under this high temperature, the components easily fell off.


Simply clean the solder pad with solder wick and we will be ready to re-flow it back.



Now it is ready for the oven again. I only re-flowed the right side of the board the same way as I did to knock off the bad-soldered components. After about 4 minutes, here is what I got from the oven:


And the final test showed that all of the three buck converter chips warmed up after several minutes of dummy load test. So I successfully fixed (remove & re-solder) the board by partially re-flowing the PCB.

Defining copper to drill hole clearance rule in Altium Designer

The “copper to drill hole” distance is an important rule for circuit board design, just like the “copper to copper” clearance. A PCB manufacturer usually announce their minimum allowable clearance between copper and drill hole as part of their capability. For example, the PCB house that I am always using has a minimum 6mill for copper to copper clearance, and 10mil for copper to drill hole. This rule sometimes doesn’t need to be taken care of if the holes come with annular ring (e.g. via hols or through-type pads), because the copper to copper clearance rule (e.g. 6mil) + the width of annular ring (e.g. 5mil) is usually more than enough to satisfy the copper to drill hole clearance requirement. However, when you only need a bare through hole on your PCB board for mechanical purpose such as mounting a connector, the copper to copper distance (e.g. 6mil) is used to monitor the clearance (e.g. require at least 11mil between a hole and copper) when there is no annular ring “protecting” the hole, which may potentially cause problems.

In this article, I would like to talk about how to set up the rule for the “copper to drill hole” clearance in Altium Designer. I wouldn’t have to write this if it were as easy as the “copper to copper” rule. For some reason, Altium Design doesn’t specifically provide a default methods in the “Design” -> “Rules” menu, where you can easily define the rules such as copper to copper distance. However, Altium Designer provides a powerful “Advanced (Query)” method with which you define the clearance rule between A and B. Apparently,  as you can see from the following plot: A stands for the bare through hole (I call it “NPTH” for Non-plated Through Hole), and B for all other coppers. So just put “(InPadClass(‘NPTH’))” in the top “Full Query” windows and “All” in the bottom.


Next, you will need to define which holes are “NPTH”.  In “Design” -> “Classes”, you can manually setup a new class under the tag “Pad Classes” and name it “NPTH”. Click the NPTH and pick up the holes that you want to put under the “copper to drill hole” rule from the “Non-Members” list and add them to the “Members”.


A useful tip for easily selecting the holes: Assign “-1”  to the designator of the non-plated through hole so that you can easily filter them out with regular expression “*–1” in the “Non-Members” list. Otherwise, you just get lost in the long “Non-Members” list of all the pads in your design.

Now, if your copper gets too close to the hole, the rule monitor will give you a warn immediately!

Reflow oven soldering without stencils

First of all, let’s take a look at the video clip below (link) to find out how it is possible to reflow solder components without stencils.

I talked about DIY your own reflow oven in my previous article. People may wonder that only a functioning reflow oven won’t do the job as you must have stencil to dispense solder paste. Yes, stencils are usually considered as key piece of equipments for reflow soldering process. The purpose of stencil is to restrain the solder paste within the specific pad area. The thickness of stencil also decides the amount of the paste to apply for soldering. Stencil, however, is non-trivial to manufacture. It usually requires special material and laser-cutting. That’s why stencil is quite costly, sometimes even more expensive than your PCB board.

With_Stencil Without_stencil

Figure 1. Left: Solder paste applied with stencil; Right: Solder paste applied manually (without stencil)

The difference is quite obvious: you get a clean and neat solder paste dispensing with the help of stencil. On the contrary, the PCB on the right side looks messy as the paste is simply dropped on to the pads without a stencil. Messiness might be acceptable considering this is not your final product. The “bridge” where paste (usually applied too much) connects to its neighbors across different pads seems to be a problem because it might cause short-circuit.


Figure 2. The sketch illustrates how paste is bridging across two pads.

With the help of “surface tension”,  the PCB on the right side of figure 1, however, is still a useful board for reflow process. As we know when solder paste is heated to certain temperature, it melts and becomes “liquid”, which will flow in compliance with surface tension. The molten solder that was placed in between pads (also called solder mask) will be pulled away by its neighboring paste that was placed right on the pads. Further more, surface tension also helps the alignment of components. If a component is not accurately positioned over its footprint, the surface tension will help “move” it into place when the solder paste melts. The video above contains a clip towards the end shows how small SMT LEDs are “pushed” and “rotated” into place when the solder paste melts.

So, as long as you have a reflow oven, chances are that you can start soldering your small footprint ICs without stencils or automatic mounting machines!

Convert a toaster oven into a PCB reflow oven (video)

My reflow oven in action!

Although I can skillfully solder as small as 0.5mm pitch QFN package chip by hand (post, video), more and more space saving components appearing on the market are totally IMPOSSIBLE to hand solder. As my project advances, I have come to an point where my solder iron is becoming a piece of useless tool. So I am thinking of bringing in a reflow oven to my workbench. There are quite a few options on the market, either professional or hobbist’s. But nothing is more fun than building a reflow oven by myself, not to mention it also saves quite some money.

To build a PCB reflow oven for my workbench, I need four main components: 1) Heating device; 2) Temperature sensor; 3) Controller; and 4) Switch. Fortunately, they are not too hard to have for an electrical engineer.


Here is my choice:

  1. West Bend 74206 Convection Oven from Amazon. The first consideration for a proper oven is its power. It must be able to heat up to at least 500°F or 260°C in a short time. Usually a 1000W oven will work for this requirement. Next, it must be a convection oven because a convection oven comes with a fan that circulates the air and helps heat up every corner of the circuit board evenly. The configuration of heating elements should be 2 on the top and 2 at the bottom. Also, it doesn’t have to be large. A larger oven usually heats up slower because of its big inner space. West Bend 74206 turns out to be a little bit too big, and couldn’t raise the temperature to 150C in 90 seconds. I would prefer a smaller size if I bought another one. Overall, it works pretty well. Last but not least, I strongly recommend a mechanical switch mode rather than fancy digital control panel for easier hacking.
  2. A TPI GK11M thermocouple and a thermocouple conditioner IC MAX6675 from Digikey.  Make sure your thermocouple works under your desired temperature range. For example, GK11M has a range of -40 ~ 950°F (-40 ~ 510°C). Don’t use any semiconductor-based temperature sensor because they only function under 80°C. MAX6675 is Cold-Junction-Compensated K-Thermocouple-to-Digital Converter that has micro-controller-friendly SPI interface. Since this chip does almost everything for you, it is a little pricy. You can also choose analog output type of converter for only several dollars, but you have to implement your own ADC.
  3. I am using my own PIC32-base RTM microcontroller platform that run at 80MHz. Data can be transferred in and out through its USB2.0 port with little programming required.
  4. Crydom’s PF240D25 solid state power relay from Digikey. Since I am going to switch on/off AC power, a power relay that is capable of deliver 1000W+ @ 110VAC/220VAC is a must. If you plan to do PWM, a solid state relay is preferred than a mechanical one. Solid state relay PF240D25 can be controlled directly from PIC output, which is a plus.

After all the components have arrived, it is time to hack the toaster oven into a PCB reflow oven!


Figure 1. Notice that the toast oven came with three switch wheels (in green circles). We need to disable the top temperature limiter (which is actually a bimetallic strip) because we want to control the temperature with our own circuits. “Function” switch is not really needed here, but we can leave it intact in case of some low temperature application where only part of the heating rods are needed. Although “Timer” is also not required, I strongly recommend keep it work in circuit just in case people forget to turn it off.


Figure 2. On the inner side of the oven’s control panel. The one with big coil wrapped in white tape is the convection fan. It plays a critical role to cycle the air when it is being heated, so that the temperature will be even across your PCB board. Make sure your oven has the fan. As you can see, the wiring on the top wheel are disconnected. A small modification is also made on the fan to make it work all the time as long as the power cord is plugged in (Usually the “bake” mode will disable the fan) .


Figure 3. Rewiring is finished! As soon as the two blue connectors are connected, the oven will start heating up. Since we didn’t find any power management inside the oven, the cables are apparently running AC power directly. This makes my life easier as I only need to find a relay that can switch on/off AC by software. On the other hand, you should pay attention not to get executed!


Figure 4. According to its labels on the back, the solid state relay is quite straightforward for wiring.  Pin 1/2 connects to the blue terminators in the previous photo, doesn’t matter which goes to which. Pin 3 should be connected to the control output of micro-controller, it doesn’t seem to need a current limit resistor in series for pin 3; And pin 4 is ground. Special attention should be paid not to overheat the relay as the solid state relay is semiconductor based device, so it generates a lot of heat during operation. Extra cooling is recommended.


Figure 5. With everything connected, my PCB reflow oven is almost ready to heat up! Ohh…wait….Where is the cooling fan for the relay?


Figure 6. Now the hardware is ready! I implemented a PID loop for the temperature control to follow the required profile. There are hundreds of profile available from different sources, most of which are divided into three different sections: 1) Pre-heat zone; 2) Soaking zone; and 3) Reflow zone;

  • Pre-heat requires the temperature reaching 150°C with 2 minutes. 150°C is below solder’s melting point, just to make sure the whole board is heated up.
  • Soaking zone requires the temperature staying around 183°C for no more than 2 minutes, activating the soldering agent, evaporating necessary liquid, but still not melting the solder paste.
  • And finally, the reflow zone raises the temperature to 250°C as soon as possible. Components are soldered to its pads during this final stage.

After quite a few try and force, I have finally tuned my PID parameters to follow the desired curve. The command I send to my RTM controller is as follows:

120, 150
240, 183
300, 250

Very simple, the first column stands for time(second), and the second column is the target temperature when time is out. And the following plot is the real-time plot of my oven’s performance. As I mentioned before, the oven is a little too big too heat up as fast as possible. I wasn’t able to reach the desired temperature on time. And the peak reflow temperature didn’t soar up to 250°C either.

Amazingly, the oven works perfectly well. Check out some photos of baked PCB boards. New photos will be added after I have tried more.

Boards I have reflowed in my oven:


Reflow of 1.7mm x 1.3mm Rebel Z series 500mA powerful color LED.


Reflow of Rebel Z colorful LED and 3mm x 4.5mm Rebel white LED. The rebel Z series provides a rich selection of colors: Green, Blue, Red, Orange Red, Royal Blue, Amber, Deep Red and Cyan.

Added on June 21st, 2014:

It’s been more than year since I converted this oven into a reflow. It is quickly becoming a critical piece of equipment on my workbench. I am boldly choosing small footprint components for new circuit design because I have the capability of prototyping them on my own.

Here are some photos of a recent project that utilized this reflow oven:

Apply solder paste on laser-cut stencil
Apply solder paste on laser-cut stencil
Solder paste stays well on the pads
Solder paste stays well on the pads
Components are in position.
Components are in position.
Ready for reflow!
Ready for reflow!
Check out the results!
Check out the results!