Keith's Electronics Blog

What’s the difference between these two boards?

• The one on the right is prettier.
• The one on the left has the driver chip soldered in the correct orientation.
• The one on the left still has the magic smoke.

Tinnit is a tin-plating solution for coating copper PC boards. It dissolves in water and plates boards without electricity (which standard electroplating requires). The silvery coating prevents the copper from oxidizing, and also takes solder more easily than even clean, bare copper.

I’ve had an unopened package in my PCB-making kit for about eighteen years now. The problem is, once mixed, it has only a six-month shelf life, and I’ve never expected to make enough boards within six months to be worth using up the package.

Yesterday I finally took the plunge, and I’m rather impressed. Everything I did all day worked out pretty well . . . so let me get to the Tinnit in my usual roundabout way.

Stepper Driver on 24-30V

I borrowed a DC power supply from the class lab to test my stepper driver on 24V and 30V load supply. I had got up to 2300 steps per second on 18V, and I hit 2700 steps per second at 24V and ran out of speed on my controller. I’m not using the Arduino’s PWM outputs yet, and 2700 steps / second seems to be about as fast as my program goes using software-based stepping.

I’ve been running the motor in half-step mode because it’s noticeably smoother, at least at lower speeds. But that also means I only get half the motion for each step signal I send. For the sake of testing top speed, I rejumpered the driver board for full-stepping and tried again.

The motor was immediately jerky and twitchy, like it had been with inductive sense resistors. Not only was the top speed drastically lower, but it took great care to ramp up to the top attainable speed. I jumpered back to half-stepping — and the problem didn’t go away. I checked all my connections, measured parts of the circuit, swapped the motor, put a scope on the sense resistor terminals — no go.

I broke it. Still don’t know how.

I’d been planning to build another driver board soon anyway, to drive a second axis prototype. With my first board misbehaving, I could also use the second board to help troubleshoot the first.

I also had a list of changes to make — things that were inconvenient about my first A3977 breakout board. I updated my layout Saturday with (among other things) nice jumpers for option settings, better grouping of signal inputs and outputs, and less reliance on having to solder leads on both top and bottom sides to carry signals through unplated vias.

Hand-Drilling and Etching the PCB

Normally I’d use Joel’s CNC machine to drill my board, but the timing didn’t work out this weekend. Instead, to prepare for hand-drilling, I taped together two sheets of my thin double-sided board stock with a printout of the top-side copper pattern and used a center punch to mark the holes.

Alas, I was a bit overzealous with the center punch.

The dimples went through both boards in the stack. Yeah, thin board stock.

I drilled the holes carefully with a #71 bit, then used a hammer to flatten the dimples on the anvil surface of my bench vise. Apparently I did a good enough job, because I made a nearly flawless two-sided iron-on transfer on glossy paper (not even bothering with Press ‘n’ Peel Blue), using the paper towel between the iron and the board again to distribute the pressure.

The board came out as nicely as any I’ve done. I used 16 mil traces and 40 mil isolation, so it wasn’t particularly challenging. Still, it etched very cleanly.

I did have a little trouble with pockmarks in some of the large copper areas. They don’t impact the performance — just the appearance. I could perhaps have sanded and polished the board to make it prettier; maybe next time.

Tinnit

On each of the last two PC boards I made, after partially populating the board, I really wished I had tried using the Tinnit. And I’m making more boards now — I know I’ll make at least one more stepper driver board (for the third axis), and probably lots of other things, in the next six months. So it seemed like a good time to try it.

The instructions say to mix the Tinnit powder package with 12 oz of hot water, then add enough water to make a pint. To plate a board, heat the solution to 120-140°F in a pyrex container, then dunk the board for 35-40 minutes, agitating every five. Store the leftover solution in a metal-free container.

Well, my package had two smaller powder packets in it, and it wasn’t clear whether that was two doses or two parts of a single dose. Nor could I find any help online. I took a chance and poured in both packets; and when the second packet caused it to start foaming, it was clear that they weren’t the same and both packets were in fact required.

I mixed the solution in a glass candy jar that I got from a vendor about eight years ago. The jar was just big enough to dunk yesterday’s PC board, so I sidestepped a search for an appropriate separate pyrex container for heating the solution. I put the whole jar (minus the lid) onto a stove burner turned to low and checked the temperature and stirred frequently (with a plastic spoon). It’s not obvious from this picture, but there’s a flaky white precipitate at the bottom of the jar that only went fully into solution at about 130°F.

The board started turning silvery almost immediately, and coverage didn’t change visibly from about ten minutes in solution until forty minutes when I took it out. I washed and scrubbed it pretty thoroughly, and it still smells like salts.

The result is very shiny and silvery. The pockmarks are still visible — Tinnit (and electroplating in general) doesn’t cover to enough depth to smooth a rough surface — but the board looks pretty good.

I’m particularly impressed by how much of a difference it makes in soldering. I know it’s easier to solder tinned components, but clean copper takes solder reasonably well. Still, this was amazingly easier to solder. When I built my first A3977 breakout board, I was flowing solder across multiple PLCC leads and cleaning up solder bridges; last night, I was able to solder individual leads with my SMT soldering tip.

As I was stuffing jumpers and headers, I noticed how monochromatic the board was without the resistors and capacitors installed yet. I finished all the black, silver, and white components and stopped for the evening. Tonight I expect to add the passives and hopefully get the board up and running.

According to the explanation in my previous post, the Allegro A3977 stepper motor driver should be able to run motors faster with a higher load supply voltage. Naturally, I was eager to see this for myself.

Yes, yes it does.

Throwing Together an Unregulated 18V Power Supply

I’d been using my benchtop PC power supply for both 5V logic and 12V load supplies. After swapping the sense resistors this weekend and improving my step rate from 600 steps per second to about 1450, I was really curious how much of an additional improvement I’d get from a higher load supply. The 3977 is rated for a 35V load supply, but some forum posters recommend going no higher than 30V, so I was looking to try out 18-24V.

Joel just dismantled a photocopier with 20V motors; but after looking at the power supply boards, it wasn’t obvious which of the several power supplies was the one providing 20V. Plus they were switching supplies, and I don’t know whether they had integral loads or whether I’d have to provide a minimum load externally . . . bluck.

So I hacked together my own ~18V supply.

From left to middle to top, that’s a brand new Radio Shack 12.6V 3A transformer, 4A bridge rectifier from a dead PC power supply, and a 18000μF capacitor out of a workhorse Accel printer from a former employer. The line cord is from a broken coffee grinder, the five-pin socket came with the capacitor and the plug from Slim, and the remaining wires and connector are from the PC power supply.

Side note: I’ve seen a few big capacitors in my day, but these impress me. When servicing these printers, I once vaporized a chunk out of my pocketknife’s screwdriver blade when I discharged the big capacitor by accident. I know personally that the last time this capacitor’s printer was powered on was no more recent than 1998. On a lark, after getting the capacitor out of the printer, I put a voltmeter across it, and it still had 3V on it. Ran an LED for a couple of minutes while I was draining it. I screwed a 1KΩ resistor across it to drain it a little faster in my power supply, so I don’t have to worry as much about hurting myself.

Power Supply Calculations

I have never kept the formulas for different rectifiers in my head, so I’ll run through the numbers for this one here. The transformer’s secondary (V₂) is 12.6VAC RMS, and the peak voltage of a sine wave is RMS * √2, so the secondary peak voltage (V_P2) is

V_P2 = V₂ * √2 = 12.6V * √2 ≈ 17.8VDC

The peak DC output of a bridge rectifier (V_P) is the peak secondary voltage minus the two diode drops, so

V_P = V_P2 – 1.4V = 17.8V – 1.4V = 16.4V

The output of a capacitor input filter is the same as the peak DC value of its input, minus load-dependent ripple, so I should be getting about 16.4VDC out of this setup. Measured about 17.1, so I’m in range. The transformer is rated for 12.6VAC output at 3A, so it’s almost certainly putting out a little higher voltage with such a light load. If I weren’t so lazy, I’d go measure it, but that’s out of the question.

And there’s no regulator. Linear regulators convert electricity into heat, and I need neither less electricity nor more heat. The A3977 is effectively a switchmode regulator for the motor coils anyway; so by regulating its input load supply voltage too carefully, I’d just be duplicating effort. Especially for a test. Note that I’m still running the logic supply voltage from the well-regulated 5.0V output of my PC power supply — the red and black wires in the photo above are picking that up and adding it into the four-pin supply jack.

With my scope across the capacitor terminals, I estimated .2V peak-to-peak ripple when running one motor at full speed. That’s pretty lousy ripple for most digital or analog circuits, but by all appearances just fine for this application. Especially, as I intend to keep pointing out, for a test.

And Yes, the Motor Runs Faster

The motor is running noticeably faster before losing steps — up from 1450 steps per second to about 2300 steps per second. That’s an improvement from 2.8 seconds per inch of linear travel to 1.7 seconds per inch — a considerable difference for something as simple as changing the load supply voltage.

I want more, of course. I still want to see what happens at 24V and at 30V, so I bought a bargain 24V power supply on eBay today. When it gets here, I’ll give it a shot.

It’s worth noting that as far as I can tell, I don’t need to change anything in the A3977 circuit to adapt to the new load supply voltage. The sense resistors determine the maximum current that should flow through the motors, and the whole job of the 3977 is to keep that the same regardless of supply voltage. And I provide V_REF from a divider on the logic supply voltage, so that’s also independent of load supply. Slick.

The stepper motors I’m using to build my CNC machine are rated for 2.5A. Now, when they’re holding one position, it’s easy to deliver 2.5A of current. But when the motors are rotating, their coils are being energized one at a time in sequence.

The coils are inductive, and inductors resist changes to current. So when a coil is first energized, it takes a moment for the current to build up to capacity. As a result, with a simple stepper driver, the faster you turn the motor, the less of the time each coil gets full current, and the weaker the motor gets.

The Allegro Microsystems A3977 stepper driver chip I’m using to control my stepper motors overcomes this by running at a (much) higher supply voltage than needed to deliver 2.5A, and by using sense resistors in series with the motor coils. When it’s time to energize a coil, the A3977 delivers the full supply voltage and watches the sense resistor. When the desired amount of current is flowing through the resistor (hence the motor coil), the A3977 shuts off the voltage (briefly), then ramps it up again. The result is the delivery of the desired current on average, but with a much more rapid start due to the higher supply voltage.

Buried within Allegro’s A3977 FAQ is the recommendation that the sense resistors should be non-inductive. Since the values required are the sub-ohm range, commonly-available resistors are in big sand packages — and as noted by Jeff Mann in a comment posted about my control board, are internally wirewound and inductive.

I had already bought Ohmite non-inductive sense resistors from DigiKey, which are also much smaller and more manageable than the big sand resistors. (The cross-section of the sand resistors is about 1/2″ square.) I hadn’t worked on the stepper driver in quite a while, though, and I hadn’t put in the noninductive resistors yet.

Did that Saturday. The difference in performance is amazing.

Performance Gains

The last time I did a speed test of my stepper motor controller, I was able to get about 500-600 steps per second before the motor started losing steps and stopped turning. Additionally, it growled and twitched when it was idle, and grunted and jerked at low speeds, smoothing out only at higher speeds.

With the non-inductive resistors installed, all the growling and twitching went away. (Someone should look at non-inductive resistors for bulldogs, but I digress.) I’m able to ramp the motor up smoothly from a stop to full speed.

And full speed has increased from about 600 steps per second to over 1450 steps per second, just from installing non-inductive sense resistors. With my 200 step/revolution motors and 20 tpi all-thread, that’s an improvement from about seven seconds per inch of linear travel to about three seconds per inch.

That’s still running on 12V out of my spare PC power supply, and I’m trying to rustle up a 24V supply to test with, which should increase the speed further. That’s also without tweaking the A3977′s component values — I hear there are some tricks to play there, too.

Executive summary:

Inductive sense resistors with the A3977 make the motor growl and jerk when idle and at low speeds, and limit maximum speed. Non-inductive sense resistors with the A3977 make the motor smooth and quiet at all operational speeds, and dramatically increase maximum speed.

Z Axis Completely Assembled

I have the drill sled mounted in the frame, the motor properly mounted, and the joystick controlling it at variable speed. Next step is building the Y axis around the Z frame.

Credit Where Due

Many thanks to Jeff Mann for his original comment, which drew my attention back to replacing the sense resistors, and for our followup conversation, which helped me understand and think through how the PWM driver makes the motors work better at higher voltage without going over the rated current.