Category: Electronics Workbench

Electrical & Electronic gadgets

Arduino Leonardo + TC4 Thermocouple Shield + TC4 Server

Because I want to measure several fairly high temperatures in and around the extruder with the LinuxCNC controller for the M2, I need a multi-channel thermocouple input. LinuxCNC’s hal_input interface module exposes the values produced by USB HID peripherals as HAL pins, which seems to be a nice way to add devices. A bit of searching revealed the TC4 Arduino Shield four-channel thermocouple + four-channel thermistor + assorted I/O board produced by the homeroasters.org folks as part of their DIY coffee roast controller.

The gotcha: ordinary Arduino boards cannot (without extraordinary effort) become USB HID peripherals, as their USB interface works only as a serial data device. The solution: the new-ish Arduino Leonardo has an on-chip USB interface that can act as a USB HID peripheral as well as the usual serial device.

Dan Newman, of Sailfish firmware fame, conjured up an a Arduino program (which, IMO, is far more complex than a mere sketch) that provides both a human-readable terminal interface and a LinuxCNC HAL-compatible USB HID interface using the Leonardo’s USB capabilities; the TC4Server project repository is on my GitHub account to keep it close to all the other stuff that should appear over the course of this project. His firmware builds on the libraries for the homeroaster’s tc4-shield Google Code project, but is intended for use with LinuxCNC, rather than as part of an Arduino controller.

It’s worth noting that the Leonardo has a mere 32 kB of program storage, so the extensive help documentation built into the program helped prevent feature creep.

Although I’m not yet using LinuxCNC with the M2, I can use TC4-Server’s serial terminal interface to read four channels of thermocouple data to help figure out what’s going on with the M2’s extruder thermistor. The TC4 shield has screw terminals for the thermocouples, but I also added a Proto ScrewShield board for thermistor resistors and easier connections:

TC4 on ProtoScrewShield on Leonardo

The TC4 Shield PCB layout assumes it’s being used with the original Arduino series of boards that bring the I²C (aka I2C, IIC, etc) SDA and SCL signals out on the A4 and A5 analog input pins, respectively. The newer Leonardo board brings SDA and SCL out in a separate header, so you must hotwire them across the board. The green and blue wires (stripped from a ribbon cable) accomplish that purpose: they’re plugged into the new Leonardo header through bent male header pins and clamped into the ScrewShield terminals. This assumes the Leonardo’s A4 and A5 pins remain as inputs, which is true for Dan’s firmware. If you actually need those pins for analog inputs, then you must remove the header pins that interconnect the boards and hotwire directly to the TC4 headers.

The TC4 shield includes an on-board temperature sensor that serves as the thermocouple cold-junction compensation reference. In my simple tests, the board has about 1 °C of self-heating, so I also use it to report the ambient temperature.

With all that in hand, I connected the thermocouple epoxied to the M2’s nozzle to the TC4 and re-ran the previous test with the modified thermistor table:

Rescaled extruder – thermocouple in TC4 board

The TC4 shield produces the same result as the Fluke meter with the same thermocouple, so we know those results will be consistent.

The modified thermistor table produces results that overlay thermocouple data. The questions come down to the accuracy of the thermocouple and whether bending the thermistor table actually represents an inherent property of the thermistor or just compensates for another problem.

Part of my motivation for using thermocouples, rather than thermistors, is that thermocouples avoid the whole dance of matching a given thermistor with a set of properties. Given the uncertain provenance of most thermistors, I have no reason to believe any of them match their alleged datasheets…

2013-05-16

Makergear M2: Thermistor Tables

The Marlin firmware used by the M2 has these thermistor selections in Configuration.h:

//// Temperature sensor settings:
// -2 is thermocouple with MAX6675 (only for sensor 0)
// -1 is thermocouple with AD595
// 0 is not used
// 1 is 100k thermistor
// 2 is 200k thermistor
// 3 is mendel-parts thermistor
// 4 is 10k thermistor !! do not use it for a hotend. It gives bad resolution at high temp. !!
// 5 is ParCan supplied 104GT-2 100K
// 6 is EPCOS 100k
// 7 is 100k Honeywell thermistor 135-104LAG-J01

#define TEMP_SENSOR_0 1
#define TEMP_SENSOR_1 1
#define TEMP_SENSOR_2 0
#define TEMP_SENSOR_BED 1

The first table in thermistortables.h looks like this:

#define OVERSAMPLENR 16

#if (THERMISTORHEATER_0 == 1) || (THERMISTORHEATER_1 == 1)  || (THERMISTORHEATER_2 == 1) || (THERMISTORBED == 1) //100k bed thermistor

const short temptable_1[][2] PROGMEM = {
{       23*OVERSAMPLENR ,       300     },
{       25*OVERSAMPLENR ,       295     },
{       27*OVERSAMPLENR ,       290     },
{       28*OVERSAMPLENR ,       285     },
{       31*OVERSAMPLENR ,       280     },
{       33*OVERSAMPLENR ,       275     },
{       35*OVERSAMPLENR ,       270     },
{       38*OVERSAMPLENR ,       265     },
{       41*OVERSAMPLENR ,       260     },
{       44*OVERSAMPLENR ,       255     },
{       48*OVERSAMPLENR ,       250     },
{       52*OVERSAMPLENR ,       245     },
{       56*OVERSAMPLENR ,       240     },
{       61*OVERSAMPLENR ,       235     },
{       66*OVERSAMPLENR ,       230     },
{       71*OVERSAMPLENR ,       225     },
{       78*OVERSAMPLENR ,       220     },
{       84*OVERSAMPLENR ,       215     },
{       92*OVERSAMPLENR ,       210     },
{       100*OVERSAMPLENR        ,       205     },
{       109*OVERSAMPLENR        ,       200     },
{       120*OVERSAMPLENR        ,       195     },
{       131*OVERSAMPLENR        ,       190     },
{       143*OVERSAMPLENR        ,       185     },
{       156*OVERSAMPLENR        ,       180     },
{       171*OVERSAMPLENR        ,       175     },
{       187*OVERSAMPLENR        ,       170     },
{       205*OVERSAMPLENR        ,       165     },
{       224*OVERSAMPLENR        ,       160     },
{       245*OVERSAMPLENR        ,       155     },
{       268*OVERSAMPLENR        ,       150     },
{       293*OVERSAMPLENR        ,       145     },
{       320*OVERSAMPLENR        ,       140     },
{       348*OVERSAMPLENR        ,       135     },
{       379*OVERSAMPLENR        ,       130     },
{       411*OVERSAMPLENR        ,       125     },
{       445*OVERSAMPLENR        ,       120     },
{       480*OVERSAMPLENR        ,       115     },
{       516*OVERSAMPLENR        ,       110     },
{       553*OVERSAMPLENR        ,       105     },
{       591*OVERSAMPLENR        ,       100     },
{       628*OVERSAMPLENR        ,       95      },
{       665*OVERSAMPLENR        ,       90      },
{       702*OVERSAMPLENR        ,       85      },
{       737*OVERSAMPLENR        ,       80      },
{       770*OVERSAMPLENR        ,       75      },
{       801*OVERSAMPLENR        ,       70      },
{       830*OVERSAMPLENR        ,       65      },
{       857*OVERSAMPLENR        ,       60      },
{       881*OVERSAMPLENR        ,       55      },
{       903*OVERSAMPLENR        ,       50      },
{       922*OVERSAMPLENR        ,       45      },
{       939*OVERSAMPLENR        ,       40      },
{       954*OVERSAMPLENR        ,       35      },
{       966*OVERSAMPLENR        ,       30      },
{       977*OVERSAMPLENR        ,       25      },
{       985*OVERSAMPLENR        ,       20      },
{       993*OVERSAMPLENR        ,       15      },
{       999*OVERSAMPLENR        ,       10      },
{       1004*OVERSAMPLENR       ,       5       },
{       1008*OVERSAMPLENR       ,       0       } //safety
};
#endif

The OVERSAMPLENR constant determines the number of successive ADC samples added together into a single value, which is then used to search the table for the corresponding entry. The table entries are pairs of:
{nominal ADC value * number of samples, temperature in C}
which means that if we know the temperature, we can work backwards to find the ADC value and then compute the actual thermistor resistance.

However, before doing that, I created a modified version of the thermistor table that simply scales the temperatures down by 0.878:

#if (THERMISTORHEATER_0 == 8) || (THERMISTORHEATER_1 == 8) || (THERMISTORHEATER_2 == 8) || (THERMISTORBED == 8) // M2 thermistors on RAMBO
const short temptable_8[][2] PROGMEM = {
	{23*OVERSAMPLENR, 263.51},
	{25*OVERSAMPLENR, 259.12},
	{27*OVERSAMPLENR, 254.73},
	{28*OVERSAMPLENR, 250.34},
	{31*OVERSAMPLENR, 245.94},
	{33*OVERSAMPLENR, 241.55},
	{35*OVERSAMPLENR, 237.16},
	{38*OVERSAMPLENR, 232.77},
	{41*OVERSAMPLENR, 228.38},
	{44*OVERSAMPLENR, 223.98},
	{48*OVERSAMPLENR, 219.59},
	{52*OVERSAMPLENR, 215.2},
	{56*OVERSAMPLENR, 210.81},
	{61*OVERSAMPLENR, 206.42},
	{66*OVERSAMPLENR, 202.03},
	{71*OVERSAMPLENR, 197.63},
	{78*OVERSAMPLENR, 193.24},
	{84*OVERSAMPLENR, 188.85},
	{92*OVERSAMPLENR, 184.46},
	{100*OVERSAMPLENR, 180.07},
	{109*OVERSAMPLENR, 175.67},
	{120*OVERSAMPLENR, 171.28},
	{131*OVERSAMPLENR, 166.89},
	{143*OVERSAMPLENR, 162.5},
	{156*OVERSAMPLENR, 158.11},
	{171*OVERSAMPLENR, 153.71},
	{187*OVERSAMPLENR, 149.32},
	{205*OVERSAMPLENR, 144.93},
	{224*OVERSAMPLENR, 140.54},
	{245*OVERSAMPLENR, 136.15},
	{268*OVERSAMPLENR, 131.76},
	{293*OVERSAMPLENR, 127.36},
	{320*OVERSAMPLENR, 122.97},
	{348*OVERSAMPLENR, 118.58},
	{379*OVERSAMPLENR, 114.19},
	{411*OVERSAMPLENR, 109.8},
	{445*OVERSAMPLENR, 105.4},
	{480*OVERSAMPLENR, 101.01},
	{516*OVERSAMPLENR, 96.62},
	{553*OVERSAMPLENR, 92.23},
	{591*OVERSAMPLENR, 87.84},
	{628*OVERSAMPLENR, 83.45},
	{665*OVERSAMPLENR, 79.05},
	{702*OVERSAMPLENR, 74.66},
	{737*OVERSAMPLENR, 70.27},
	{770*OVERSAMPLENR, 65.88},
	{801*OVERSAMPLENR, 61.49},
	{830*OVERSAMPLENR, 57.09},
	{857*OVERSAMPLENR, 52.7},
	{881*OVERSAMPLENR, 48.31},
	{903*OVERSAMPLENR, 43.92},
	{922*OVERSAMPLENR, 39.53},
	{939*OVERSAMPLENR, 35.13},
	{954*OVERSAMPLENR, 30.74},
	{966*OVERSAMPLENR, 26.35},
	{977*OVERSAMPLENR, 21.96},
	{985*OVERSAMPLENR, 17.57},
	{993*OVERSAMPLENR, 13.18},
	{999*OVERSAMPLENR, 8.78},
	{1004*OVERSAMPLENR, 4.39},
	{1008*OVERSAMPLENR, 0}
};
#endif

I inserted that table, changed the thermistor selection, reloaded the firmware, and ran the same test as before, which produced this result:

The stock thermistor and the thermocouple now report essentially the same values, which is entirely due to the new table. The two additional lines come from two more thermocouples taped to the nozzle and dangling downward toward the platform:

M2 - Hot end with additional thermocouples — M2 – Hot end with additional thermocouples

Given that I simply taped those thermistors in place, they don’t contact the nozzle nearly as well as the epoxied sensors. The fact that one reads a bit higher and the other much lower could be explained by handwaving, but one possibility is that the various thermocouples don’t quite agree with each other.

Time for some calibration along those lines, methinks…

2013-05-14

Makergear M2: Thermistor vs. Thermocouple
With the stock thermistor and my added thermocouple epoxied to the M2’s nozzle, I stepped the temperature upward, let it settle, and recorded the temperature from the Pronterface status display and my Fluke thermocouple meter:

First Heat – M2 thermistor – Fluke with thermocouple

Because the firmware servos the temperature through the stock thermistor, that line is dead straight at the exact setpoint values: the reference never disagrees with itself. The thermocouple, however, reads low by about 12%: according to it, the nozzle runs much cooler than the thermistor value.

Huh?

Several explanations come to mind:
- The firmware is using a lookup table that doesn’t match the thermistor
- The Fluke thermocouple meter reports the wrong value
- The thermocouple junction is defective
- Despite the epoxy blob, the two sensors aren’t at the same temperature
- Something else is kaflooie
This obviously calls for more data…
2013-05-13
Zinc-air Cell Corrosion

For reasons that, alas, have little to do with normal age-related hearing degeneration, I’ve been wearing Etymōtic (that should be a long o symbol) MP-915 High-definition Electronic Earplugs (aka, Martian Ear Beetles) when I need a 6 dB boost for normal conversations. The key advantage: a price 10 dB under full-throttle hearing aids.

Anyhow, each one runs for about two weeks on a 312 zinc-air primary cell, so I picked up a batch from the usual eBay vendor. These were short-dated, which let me figure out how long after the rated shelf life they’d be good for, so I know what’s the largest batch I should buy.

One cell arrived with its air vent seal dislodged:

Corroded zinc-air cell

The cell in the middle is used, with several scratches from the contact point inside the earplug. The cell on the right has a good seal.

Assuming a good seal, the cells seem to work about as well as the long-dated fresh cells included with the earplugs.

Although I found several datasheets (Duracell Energizer Rayovac), there seems to be no way to relate the actual cell you purchase to any particular datasheet; the part numbers do not correspond to anything on the package and the nomenclature varies wildly both between manufacturers and within product lines.

2013-05-12
Makergear M2: Nozzle Temperature Sensing

The stock Makergear M2 hot end uses a 100 kΩ thermistor for temperature sensing. A wrap of Kapton tape holds it against the brass nozzle, with a stretchy fiberglass-lined tube for protective insulation and a bit of pressure. This picture shows the tape pocket around the thermistor, with my thumbnail on the left:

M2 Nozzle Thermistor – as-shipped

Peeling back the tape a bit shows the thermistor against the nozzle, with the glass bead reflecting the LED ringlight around the macro lens adapter:

M2 Nozzle thermistor – exposed

The RAMBo board in the M2 has four thermistor inputs and no thermocouple inputs, which surely drove the decision to use themistors. I want to use thermocouples with the LinuxCNC controller, because they’re more compact and happier at higher temperatures.

So I unwrapped the nozzle and lined up a thermocouple beside the thermistor:

M2 – extruder – thermistor-thermocouple

Where a dab of JB Weld firmly bonds them to the nozzle:

M2 – extruder – epoxied sensors

As nearly as I can tell, the JB Weld that I used on the Thing-O-Matic is still going strong. I think the trick is to not apply mechanical force to the bond when it’s hot; secure the leads firmly and use the epoxy only as a thermal connection. Yes, you can get fancy higher-temperature adhesives, but this seems to work well enough.

For the moment, I’m using ordinary cotton cloth secured with Kapton tape as insulation:

M2 – extruder – cotton insulation

The brown dot that looks like a bead is actually a flat stain on the nozzle.

The insulation should become something more suited for high temperatures, perhaps the ceramic fiber I used on the Thing-O-Matic, but cotton will suffice for now.

Now, wisely is it written that a man with one thermometer knows the temperature, while a man with two thermometers is never certain…

2013-05-10
Belkin F6C1500 UPS Battery Wiring Arrangement

Given the length of the battery wires inside a Belkin F6C1500 UPS, you might think any arrangement will work. Not so. The wires from the guts of the UPS must exit to the batteries exactly like this:

F6C1500 Battery wires from UPS

There’s a black wire tucked under the red wire, both of which must exit though the angled slot and run toward the front of the battery compartment.

Seen from the front, the red wire connects the positive terminal of the lower (left) battery to the negative terminal of the top (right) battery and the black wire connects the negative terminal of the lower battery to the UPS circuitry:

F6C1500 Battery interconnect wires

Trust me on this: there is no other arrangement of those wires that will simultaneously connect everything properly and fit within the case.

As for disassembly, the small tab on the left end of the case holds the front panel in place. Press that inward with a flat screwdriver, then slide the cover toward the tab. Four locking slots along the sides will disengage and you can then lift the panel off.

With that out of the way, there’s a screw hidden under the BELKIN label in the middle of the removable cover:

Belkin F6C1500 UPS Cover Screw Locations

2013-05-03
Makergear M2: Z-axis Stepper Motor

Several users have observed that the stepper motor driving the M2’s Z axis leadscrew gets very hot. I measured about 140 °F = 60 °C on the as-built motor, so I loosened the screws and raised the motor slightly:

M2 Z axis motor – raised

I eased some heatsink compound underneath by putting dabs on a slip of paper and painting it on the bottom of the motor case, lowered the Z stage to the bottom of its travel, and tightened the mounting screws:

M2 Z axis motor – added thermal compound

That reduced the temperature to about 120 °F = 50 °C, which still seemed excessive for a short-stack motor mounted on a fairly large chunk of stainless steel. The motor also sounded quite rough during homing and long manual moves, sooo … something was wrong. I bet you know where this is going, right?

Let’s start with the firmware side and determine what current the motor should be seeing.

The M2 uses a slightly modified version of the Marlin firmware running on a RAMBo 1.1b board. The basic RAMBo doc gives these equations relating the peak winding current Imax to the constant W that defines it in the firmware:

Vref = 0.8 * Imax
W = 255 * (Vref / 1.66)

Mashing those together produces this:

W = 255 * (0.8 * Imax) / 1.66

The default Z axis stepper current constant W (called Z_CURRENT in the Marlin source) is 135. The board in my M2 has R30 = 3.3 kΩ, which sets the maximum possible current to 2 A. Working the equation backwards, a Z_CURRENT = 135 will produce a peak winding current of 1.1 A.

However, a nearby comment in the source code suggests this is should be about 0.75 A. The original RAMBo board had a maximum possible current of 1.5 A, but running those numbers doesn’t agree. Another comment suggests 185 corresponds to about 1 A, which isn’t right, either. There’s nothing new about stale comments not corresponding to the actual hardware; I’ve done that myself.

With 1.1 A in hand, let’s unplug the cable and measure the winding resistance.

Not much to my surprise, the motor has 28 Ω windings. The M2 uses a 19 V supply for the steppers, so the maximum motor current works out to 19 V/28 Ω = 680 mA, but it must be less than that to allow the microstepping controller to manage the current.

It seems that Makergear is connecting a high-resistance stepper intended for a simple H-bridge drive to a high-performance microstepping controller. For some background on why that combination doesn’t work, see my analysis of the original MBI Thing-O-Matic steppers.

I thought we all agreed we weren’t going to do that any more. Maybe nobody sells a low-resistance motor-with-integral-leadscrew?

Anyhow.

The only thing to do in the short term is to reduce the peak current to a rational value around 600 mA:

74 = 255 * (0.8 * 0.6) / 166

I set it to a nice, round 75 and reloaded the firmware, which immediately made the motor hum, rather than growl, on long moves. The case temperature didn’t drop by very much, because the poor motor still dissipates about 11 W, not much less than the original 13 W. There’s only so much heat you can pull out of the case and these little motors are actually rated for maybe 5 W, tops.

The motor’s overall performance didn’t change, which is good, because it didn’t have much performance to begin with. The X and Y motors can accelerate at 9000 mm/s², but the Z motor limit is 30 mm/s²; it doesn’t really accelerate, it sort of gains momentum in a stately manner.

Next: let’s see if it really matters.

2013-04-10