The brains of Argon drives have arrived!

The microcontrollers used in the first batch of Argon drives have arrived for programming. We program these chips in house and send them to assembly house so we don’t need to make an needle bed type programming device to program mounted chips. So drives become functional right after assembly will be ready for quality control & compliance testing.

400 pcs of microcontrollers for Argons

400 pcs of microcontrollers for Argons in moisture barrier packages. The black item is a tray for 160 chips (same kind trays inside packages).

The chips will be programmed by using clam shell type IC sockets. Programming one chip takes about 5 seconds + handling.

Bare chip programming devices

Bare chip programming devices

Resetting STM32 option bytes

Without too much thinking I wrote a code to Argon’s STM32 ARM processor that write protects its flash content. After a minute I encountered a problem that I was unable to program anything new to the chip as it protects writing into memory also with a debugging tool effectively bricking the device. It took an hour to figure out how to restore the chip option bytes that determine if the flash memory is protected or not. JLink’s small utility that is supposed to restore STM32 option bytes didn’t appear to work for write protection so alternative solution was necessary to avoid replacing the chip on PCB.

After some googling (found this useful forum post) and reading chip data sheets I figured out how to do it manually with JLink Commander tool. I’m posting JLink command list here as I’m sure there are people needing this info, too.

(unprotect option bytes)
w4 0x40023c08,0x08192A3B
g
halt
w4 0x40023c08,0x4C5D6E7F
g
halt

(restore non-write protected state. one could also reset other option bytes here such as read protection or brown out reset settings by modifying the next line)

w2 0x40023C16,0xffff
g
halt

(write option bytes)
w1 0x40023C14,0xef
g
halt

I’m not sure if g and halt commands are necessary but used them just in case. This should work at least with STM32F2 series and probably also STM32F3 and STM32F4 chips too.

JLink Commander when unlocking STM32F2 write protection

JLink Commander screenshot after unlocking STM32F2 write protection

We have a winner!

We have got green light from box manufacturer from the competition candidates and I’m happy to present you the winning work from Tomasz! The submitted work shows high degree of artistic skills and creativity yielding a simple yet attractive design.

Box competition winner

Box competition winner submission. The printed version will be based on either one of the upper row designs.

I and rest of the Granite Devices are very thankful for all the participants! It was fun and turned out to be very successful competition.

Finding power limits, part 2

Long time since last tech related post! Today continued the test to find Argon drive power limits (see earlier tests here). This time I loaded drive’s internal AC to DC power supply with an old room heater while simultaneously measuring transmitted power from AC and DC lines.

The heater pulled 1.53 kW from drive HV DC bus for about 45 minutes continuously with only passive cooling of the drive. Temperatures were measured from multiple points of drive along PSU current path (components & PCB traces) to find out if anything is overheating. I’m happy to report that everything stayed relatively cool which means the drive is capable of at least 1.5 kW output continuously! The hottests parts inside were 8 Amp fuses which is easily helped by using larger fuses.

Competition finished!

I would like to thank everyone attending to the box artwork competition. We definitely got some useful material that will be printed on boxes.The winner will be announced ASAP this week.

Update

At the moment we are waiting for box manufacturer’s comments about printability of some designs candidates. The winner will be announced right after they respond!

2 days left!

There’s still plenty of time to submit your idea to the box competition. Currently submitter’s chances of winning are good as no huge flood of submits (at least yet!) have been recorded.

Meanwhile the other news in Argon project:

  • I have been mostly designing the routine test for the factory making Argons. This needs quite careful thinking because as usual, everything must be tested. No single I/O pin or internal circuit feature shall pass without being verified to be good.
  • Also the design for USB SimpleMotion V2 adapter that is needed to communicate with Argons have been finalized and placed in production

Box artwork competition!

This is the box artwork competition for Argon servo drive.

Goal

The goal is to design a printable artwork for the product package of Argon servo drive. We accept all submissions from hand sketched ideas to print-ready graphics. All submissions are treated equal (see details in Prize).

Specifications & Tips

  • Box dimensions 241 x 144 x 61 mm and top lid area is 220 x 144 mm.
  • Max paint surface coverage 20%
  • Single color (can be black or some other color). Gray scales are supported.
  • All sides can be printed
  • Avoid placing important details near edges as print alignment is not that accurate
  • File format preferably in a common vector graphics format (i.e. SVG)
  • To get a free and good SVG editor, check out Inkscape

Rules

  • Post your work to this CNCzone forum thread, or if you prefer to keep it private, email to competition at granitedevices.com
  • Submission deadline 12.4.2013 24:00 GMT
  • Any amount of submissions per person allowed
  • The submitter must grant full rights for Granite Devices Inc to use the submitted work

The winner takes it all:

  • Argon servo drive from the first production batch with designer’s signature and of course boxed in the final printed packaging
  • A 50% discount coupon for purchasing up to 2 additional Argon drives
  • In other words, get one drive for free or 3 drives for price of 1.
  • Prize total value ~800 Euros

The prize will be granted for the person who submitted the work that gets printed to the box directly or after our modifications (including sketch implementation). If it should happen that none is selected, or even ideas present in the submission are used, the prize will be granted to the overall best submission voted by Granite Devices staff.

Happy drawing and good luck!

The fine print: if we find it necessary, we reserve right to adjust & update the competition rules. If this should happen, there will be list of changes made.

Boxes

From the beginning of Argon release we plan to use a fulfillment service (a third party company with item warehousing and shipping service) to ship our products around the world. This enables faster delivery times as well as reduced shipping costs. To achieve this, items must be well packed before entering the warehouse as the service doesn’t know the specifics of how the items should be properly handled or protected inside a post parcel.

This calls a rigid box for Argon drives. Various sizes and carton types have been prototyped and the one used below seems like a good fit. The boxes should also come with some artwork and item identifier printed.

Various box prototypes

Various box prototypes

I’m thinking of throwing in a box artwork design competition. The prize would be, of course, an Argon drive or more. Leave a comment if you wish to see the competition to begin in the very next blog post!

Resolvers ain’t dead

Resolver a.k.a. synchro is a position sensor born long time before encoders. They are fully passive components consisting only coils and iron which makes them very robust in harsh conditions – including hot running servo motor under vibration, dust and dirt.

Despite being largely replaced by encoders, resolvers are still surprisingly popular even in new motors. Infineon, one of the largest microcontroller companies, even introduced a new MCU with integrated RDC (resolver to digital converter).

Resolver should be interesting also for DIY people as eBay is full of perfectly fine servomotors equipped with resolvers for bargain prices.

A resolver (stator and rotor part) and an adapter card plugged to Argon

A resolver (stator and rotor part) and an adapter card plugged to Argon

Argon supports resolvers with help of a small adapter board that hooks directly to the feedback device connector. Testing of this adapter however has to wait until final drives are received as pin-out of the Argon feedback port was modified to be more soldering friendly and the adapter in the photo is already designed for the new pin-out.

Design frozen

I’m at the point where I can’t think of any improvements to the hardware, so the order of first production run to factory will happen this week! The first patch will be split to two: smaller test run and followed by full production after all changes are verified to be good.

Last changes done

The last significant change made was to replace standard Y2 safety rated filter capacitors to every-possible-way-superior Murata surface mount Y2 capacitors. Not because design would not pass EMI compliance or there would be any problem with the standard type, but merely because I like the improvement!

Murata SMD Y2 filter capacitor

Murata SMD Y2 filter capacitor

Standard Y2 filter capacitor

Standard Y2 filter capacitor

The self resonance frequency of typical 4.7 nF leaded capacitor occurs between 20-25 MHz  while the SMT type has it around 90 MHz. This yields significantly more effective EMI noise filtering which was confirmed by oscilloscope measurements.

The new cap has also much better temperature and voltage stability characteristics. All available 4.7 nF leaded type capacitors tend to have bad temperature characteristics (curves E or F in the picture below, notice -80% change of capacitance at extreme temperatures) while capacitance of the SMT type nearly stays flat over the whole temperature range. Same kind of capacitance change does not only occur with temperature but also by voltage applied to capacitor.

Capacitor temperature characteristics (top) and impedance vs frequency curves (bottom). The self resonance frequency is the dip in the middle of curves.

Capacitor temperature characteristics (top) and impedance vs frequency curves (bottom). The self resonance frequency is at the dip seen in bottom curves.

The above actually applies to all ceramic capacitors. Luckily the temperature & voltage characteristics have standardized naming so staying away from poor types is easy. Cheap ceramic capacitors are mostly called Y5V or Z5U capacitors while good caps know names such as X7R or NP0. Argon comes only with X7R or better types.

Horrors of Y5V ceramic capacitor

The horrors of common Y5V ceramic capacitor. Take for example a 35 volt Y5V capacitor and put 24V in it – capacitance decreases 90% from the specified value while ideally it should not change at all.

X7R characteristics

X7R characteristics. The worst case capacitance change stays within +/- 15%.

Back to the topic – the only downside of the SMT Y2 filter capacitors is the 4X cost compared to the standard type. However after all the total price difference isn’t huge as there’s only 3 pcs of them in Argon.