To install the heated print bed for my reprap, I placed the PCB on top of the plywood top print plate (which is slightly larger than the PCB), clamped it in place, and then used the holes in the PCB to drill out mating holes in the top plate. I put a washer on a 16mm M3 screw, and inserted it up through the print plate, added another washer, and a M3 nut to hold it securely in place. (I actually used SAE washers, as you can probably see from the photos, and that’s okay, too.)
Next I took my PCB heatbed to Lowes and had them cut a piece of glass to the exact same size. (While I was at Lowes, I picked up some Frost King 2″ insulated pipe wrap.) The glass is about 2.2mm thick, and it was only a few dollars. I cleaned up the edges of the cut glass with some fine grit sandpaper to make it a bit safer, then I put a small piece of Kapton tape (polyimide) on each corner of the glass to provide a little protection from the screws. Using the same tape, I secured the glass to the top of the PCB.
I covered the bottom of the PCB with the Frost King insulated pipe wrap to help keep the heat going up. I had to cut out notches to make room for the screws that would hold it in place (at the corners) as well as the tops of the screws that fastened the print top plate to the print bottom plate. I added a couple bits of extra pipe wrap to the center to help support the PCB and keep it from sagging.
The PCB/glass assembly rests solidly atop the four screws — there is no “play” at all. Here are a couple of pictures. (Click for larger view.)
A couple of people have noticed this, so I thought I’d write it up. There’s nothing wrong — it’s just something to be aware of if you’re soldering up a Sanguinololu 1.3a PCB.
If you look closely at the circuit board where the FTDI chip goes, you’ll notice a tiny solder bridge between pins 25 and 26 (below the “L” in “FT232RL”). I noticed it when I first inspected the board, and I looked at the original gerber files to make sure it was part of the design. It is, in fact, part of the original design (see photo below), so there’s no need to try to remove it. (To confuse the issue, the instructions even say to look for things like this.)
If you did remove it, that’s okay, too — those pins (which both go to ground) are still connected inside of the outline of the chip.
One other thing to note is that the pin headers along the top of the board sit a little further in than those along the other sides. This is also exactly as it is in the original gerber files, but it means that the PCB might prevent some right-angle headers (like those in the Mouser project kit) from plugging all the way in unless the male headers are soldered a little high. (See the second photo.) This is probably more of an aesthetic issue than a functional one, but if you want to make sure they plug in completely, you can use one of the female connectors to space it properly when you solder it on.
Yeah, free. If you’re starting a 3D printer project (such as RepRap) at your public K-12 school in the United States for the benefit of the students, then we’ll give you a free Sanguinololu v1.3a PCB to help you out (while supplies last). They typically sell for about $12-14. We won’t even charge you for shipping.
What is a 3D printer? Check out the video at the bottom of the page for an introduction.
Not a school but want one anyway? I’m selling some of them to raise funds for a printer for a local school. Just select “None – Buy one” for affiliation below, and I’ll send you the details. They’re $11, shipping included within the USA.
Sanguinololu 1.3a PCB
We don’t see any reason why kids shouldn’t have access 3D printing technology in their problem solving toolkit, and soldering up the electronics is all part of the fun. A club with only six kids raising $50 each could be well on its way to building and operating a 3D printer for its school, and while the electrionics can typically amount to 1/3 or more of the final build price, hopefully a free PCB will help ease the pain a bit.
Are you interested? Submit the form below for instructions on how to get your free Sanguinololu 1.3a PCB.
The shipping “estimator” at Mouser.com is a little vague, so I thought I’d post the results of a recent order. In general, you’re not going to know the shipping cost or the weight of your order until after it’s shipped.
My recent order totaled $189.12 and had 33 line items, including all kids of components (resistors, capacitors, ICs, connectors, crystals, buttons, crimp terminals, etc), in quantities ranging from 2 to 250, primarily for assembling four Sanguinololu circuit boards (with some exceptions, plus some other stuff).
Selecting residential delivery via USPS to North Carolina, USA, the Mouser shipping estimator listed $6.95 for 1 pound and $9.60 for 2 pounds. I guess that the order wouldn’t weigh any more than 2 pounds. The actual invoiced shipping cost turned out to be $7.74.
Although the order was prepared the previous night, I didn’t remember to actually place it until just after 4:00 EST on January 25, 2012 — so I had just missed the same-day shipping cut-off time. The order shipped on January 26, and the credit charge (for the exact invoice amount) was posted on January 27.
The delivery tracking updates were as follows:
Out for Delivery, January 28, 2012, 9:55 am, CHAPEL HILL, NC 27514
Sorting Complete, January 28, 2012, 9:45 am, CHAPEL HILL, NC 27514
Arrival at Post Office, January 28, 2012, 5:08 am, CHAPEL HILL, NC 27514
Electronic Shipping Info Received, January 27, 2012
Depart USPS Sort Facility, January 27, 2012, FORT WORTH, TX 76161
Processed at USPS Origin Sort Facility, January 26, 2012, 8:18 pm, FORT WORTH, TX 76161
Accepted at USPS Origin Sort Facility, January 26, 2012, 7:03 pm, MANSFIELD, TX 76063
One final note: Recently I’ve noticed that there are a number of different ways people pronounce Mouser. Recently I’ve heard Mouse-er (rhymes with house-er), Moze-ure (rhymes with rose-ure), and Mau-zer (rhymes with cows-er). So how do you pronounce Mouser? It’s pronounced Mau-zer (rhymes with cows-er), but I still like to think of it as Mouse-er.
I’ve set out to build a Prusa Mendel for $300. So far I think I’m doing pretty well. The top part is what I’ve already picked up, and the bottom part is what I have yet to purchase — that’s where I still have some flexibility in cost. The first column is the percentage of total cost for that part. Shipping charges for a group of items from the same supplier are listed with the first item in that group. [Note: This chart has been updated many times as I build, and as because I'm already up and running with my reprap 3D printer, there's no longer anything left to purchase.]
Where I discovered mistakes, I’ve corrected them. For example, I actually ordered nylock #6 nuts instead of regular ones. In cases like these, I’ve just corrected the prices and pretended like I never incurred the cost of the wrong product.
There’s one important point I’d like to make out here: I was patient, I waited for good deals (and occasionally got exceptional ones), and I arranged some group buys for bulk discounts. Often if you agree to purchase larger quantities of a product, you can secure a discounted rate, and so that’s what I did where I could. You might not find all of the same deals I did, but you may get pretty close if you are patient and work with your suppliers.
I’ve added a table below that breaks out the electronics order, including what I had on hand, what I bought from ebay, and what I didn’t actually need. (Welcome Hack A Day readers!)
Finally, be sure to scroll to the right in the spreadsheet views for important notes or shortcuts on most line items.
40-pin female headers are inexpensive and easy to trim down to size, and with a dot of CA glue (super glue), you can even easily make other configurations, like 2×4. Sometimes they’re even described as “break-away” headers, although admittedly they don’t break away as easily as the male pin headers. Just keep in mind that if you don’t score it, it could shatter.
Someone was looking at one of my PCBs and commented that it’s “impossible to break them apart without destroying at least 2-3 pins each time.” Using this technique, I’ve never “destroyed” any more than one, and with a little care, the edges turn out quite nicely.
Here’s a 12-channel bidirectional logic level shifter with a 3.3V power regulator that I designed so I could easily interface 5V circuits with cool 3.3V widgets like the ADXL345 accelerometer, the NRF24L01+ wireless transceiver, and the Nokia 5110 84×48 LCD. For the one in the picture, I used round machine pin headers because I use it with hookup wire and those female headers are very easy to separate from 40-pin lengths.
DIY 12-channel Bidirectional Logic Level Shifter / Logic Level Translator, using AMS1117 and 2N7000 MOSFET pass transistor interfaces
Up to twelve 5-15V logic signals can go into one side and come out as 3.3V signals on the other side, or the other way around (3.3V signals go in and can be read as 5-15V signals) in any combination. The regulator portion converts 5 to 15V DC into a steady 3.3V for up to 1 amp.
Why did I not just use a 74HC4050 or some other logic level translator? Because this is much cooler! It’s bidirectional, it has twelve independent channels, in includes a 3.3V regulator, it’s inexpensive, it’s good soldering practice, and it’s hackable. In the photo below, I’ve wired it up to specifically drive a 3.3V Nokia 5110 LCD. It uses only five of the 12 channels as logic level translators, channel two is wired directly to 3.3V for power, and channel 3 is wired directly to ground. The transistor on channel nine is turned so that the 5V signal controls the gate, which allows PWM control of the LED with full power from the regulator. I’ve also used a different resistor. (More photos below.) And why no resistor arrays? Because resistors are usually inexpensive and on hand, and resistor arrays would limit hackability.
Logic Level Shifter Hacked for Nokia 5110 LCD
It’s essentially a AMS1117 3.3V regulator with capacitor at the top, and then an array of MOSFET pass transistor interfaces with n-channel 2N7000 MOSFETs. Each side is held high by a resistor 5.6k ohm resistor. If the 5V side is pulled low, the voltage from ground to source will be 2.7V because the diode becomes forward biased, and this turns on the transistor, pulling the 3.3V side low. If the 3.3V side is pulled low, the transistor also turns on with 3.3V, pulling the 5.5V side low.
It was designed using KiCAD (free), and PCBs were ordered through SeeedStudio’s Fusion PCB service. All the components are easily available online, at Digikey or on eBay. Here are the parts I used. The costs reflect per items costs for small bulk purchases (usually around 10-50 pieces) on eBay.
1 AMS1117 3.3V 1A Voltage Regulator – 28¢ (datasheet)
1 22uF electrolytic capacitor – 9¢
12 2N7000 n-channel MOSFET – 60¢ for 12 (datasheet)
24 5.6K ohm resistors – 43¢ for 24
2 12-pin female headers (I used round ones for the one in the picture, but you should use the standard square ones if you’re not just using jumper wires.) – 29¢ for one 40-pin header, then cut into smaller pieces.
2 2-pin female headers
1 custom PCB – $1.12
The total unit cost was $2.81, plus time and solder. I’ve decided to sell off the extras either as PCBs or as kits, so let me know if you’re interested by commenting below.
Here’s the schematic and a screenshot of the PCB design, as well as a larger version of the image above:
DIY 12-channel Bidirectional Logic Level Shifter / Logic Level Translator, using AMS1117, MOSFET pass transistor interface, and 2N7000 MOSFET transistors.
Schematic for 12-channel Bidirectional Logic Level Shifter
PCB for 12-channel Bidirectional Logic Level Shifter
Logic Level Shifter Hacked for Nokia 5110 LCD
Logic Level Translator with Nokia 5110 LCD
Logic Level Shifter with Nokia 5110 LCD Mounted
12-Channel Logic Level Shifter/Translator PCB, Top
7:52 AM – Order placed. The total charge appeared on my PayPal account six minutes later, and the order confirmation with order number arrived at the same time.
8:47 AM – I added the order number to the silkscreen layer of the PCB as requested and sent the gerber files to the email address specified about an hour after the order was placed. The design and gerbers were created with Fritzing.
6:26 PM – Received an order update email: The order has been changed to status “Processing.”
8/03/2011, 4:37 AM – Received an order update email: The order has been changed to status “In production.”
8/04/2011, 1:49 AM – Received a person email from Dwin, I assume because when I ordered I had asked them to let me know if there were any problems: “Your PCB is under processing. If the file and design meets requirements, they will be processed and shipped in 4-6 working days.”
4:33 AM – Received an order update email: The order has been changed to status “Traceable.”
4:56 AM – Received an order update email: The order has been changed to status “Shipped.” (Yay!)
9:21 PM – Received an order update email with a link for tracking the order.
8/09/2011, 4:44 PM – Tracking status: “Acceptance” at Hong Kong; Tracking status: “Origin Post is Preparing Shipment.”
8/11/2011, 10:53 AM – Tracking status: “Processed Through Sort Facility” at Hong Kong. “The item left Hong Kong for its destination.” (United States of America)
7:29 AM – Tracking status: “Arrival at Unit” at my local post office.
4:01 PM – Delivered. I actually received 12 (instead of 10), and they all look great. They were very well packaged in a small corrugated cardboard box with lots of bubble wrap.
That’s 21 days or three calendar weeks start to finish — not bad at all! Tonight, I solder!