The Smell of Molten Projects in the Morning
Ed Nisley's Blog: Shop notes, electronics, firmware, machinery, 3D printing, laser cuttery, and curiosities. Contents: 100% human thinking, 0% AI slop.
Concerted rummaging in the Basement Warehouse produced some rather old acrylic sheets:
Washing with detergent and denatured alcohol cleaned off a lot of grunge, but the yellow tint says it’s been around for a while. In fact, It Came With The House™ when we bought it three decades ago.
One sheet was a status board in an automobile machine shop:
So, yeah, that might be 70-year-old acrylic.
A cheap dual thermocouple meter, utterly devoid of branding, arrived:
It seems suitable for a semi-permanent laser cooling water monitor, particularly because it can perform arithmetic to show the difference between the inlet and outlet temperatures. The minuscule clock face at the center top of the display shows it’s in auto-power-off mode, which can be defeated by a Vulcan Nerve Pinch while turning it on.
Having a large backlit display was a selling (well, buying) point and the instructions have this to say about its operation:
The instructions say nothing about defeating the backlight timeout. The description is technically correct, because the two seconds before it goes dark is “within 30 seconds”, but I’d rather have a nicely lit display that’s on all the time.
Five screws hold the back cover in place, with no nasty prying required to pull it apart, and the build quality is about what you’d expect for a cheap meter. The circuitry fits on a single PCB and perhaps the thermistor over on the right serves as the cold junction compensation:
Doodling the backlight circuit layout suggests it’s pretty simple, even without filling in the component values:
I replaced the transistor base resistor with a somewhat larger 4.7 kΩ SMD part and added a flying wire to jam the transistor on all the time:
The IC is a serial EEPROM with its VCC and ground pins in the usual places, so, when the power to the EEPROM goes on, the backlight turns on and stays on.
The meter draws a bit over 8 mA with the backlight running, which means the trio of AAA cells won’t last all that long. When things settle down, I’ll conjure a simpleminded power supply running from a convenient voltage inside the laser cabinet.
A crude test setup to measure the duct fan’s air flow against resistance from plausible lengths of 6 inch duct and fittings:
The orange stripe (upper left corner) marks the blast gate mounted on the steel plate closing off the fireplace: when the stripe is visible, the gate is open. It’s hot-melt glued into a plywood square reducing the 8 inch hole in the plate.
I won’t be using five feet of steel duct, but [handwaving] it’s what I have on hand and should produce results similar to a shorter length of flexible duct [/handwaving].
A useful conversion factor from the anemometer’s air flow in meter/sec to the corresponding volume flow in ft³/min (colloquially CFM), based on a 6 inch diameter opening with uniform airflow:
38.6 ft³·s/m·min = 0.196 ft² × 3.28 ft/m × 60 s/min
The air flow up the chimney depends strongly on basement temperature, outdoor temperature, and wind speed. On a midwinter’s calm-but-freezing evening it ran around 1.5 m/s → 57 CFM and the next day I measured 0.7 m/s → 27 CFM with wind gusts pooting old-fireplace smell into my face.
A picture being worth a kiloword:
The upper line is the duct fan mounted as in the picture and the lower line is the bare fan as measured on the bench.
One might reasonably conclude something has gone horribly wrong, as the ductwork seems to contribute negative resistance and increased airflow. I think it’s a combination of the natural flow up the chimney, combined with a bit of flow straightening through the pipe directing air into the fan’s blades and measuring the (mostly uniform) inlet stream instead of the (somewhat segmented) outlet stream.
Anyhow, the controller has eight speeds with surprisingly linear output. I doubt the upper line’s slope of 50 CFM/click means anything, but the consistency of both suggests a 4:1 flow range, from which I can pick the lowest speed that provides enough fume extraction.
The basement has enough air leaking in (and out) that opening the exterior door had no discernible effect on the flow through the fan and up the chimney. At top speed the fan will produce two air changes per hour, chilling the basement something awful in the winter and introducing too much warm+moist air in the summer. This may call for a separate duct for outdoor makeup air, but that’s a problem for another season.
Being that type of guy, I had to measure the airflow through the inline duct fan intended for the soon-to-arrive laser cutter:
The fan is on the inlet side:
The outlet side consists of flow straightening blades around the backside of the motor mount:
The duct ports on each end are (nominal) 6 inch, with the larger central body about 7 inch ID around the blank-faced 5 inch OD motor mount.
I measured the air speed (in m/s) at the rim of the outlet port and at the center, with the rim speed about twice the center speed. The anemometer is an inch in diameter, so I assumed the annular flow was about 1.5 inch thick.
Subtracting the dead zone in the middle from the total area of the fan body gives the area of the annulus carrying most of the moving air:
| Dia inch | Area in^2 | Area ft^2 | |
| Pipe | 6 | 28 | 0.20 |
| Center | 3 | 7 | 0.05 |
| Annulus | 21 | 0.15 |
Remember, the central dead zone isn’t quite dead: it has an air speed maybe half of the annulus.
More spreadsheet action finds the flow for each of the fan speed settings:
| Speed | Outer m/s | Outer ft/m | Uniform CFM | Annulus CFM | Inner ft/min | Inner CFM | Total CFM | Rated |
| 1 | 1.8 | 354 | 70 | 52 | 177 | 9 | 61 | 44 |
| 2 | 2.9 | 571 | 112 | 84 | 286 | 14 | 98 | 88 |
| 3 | 3.8 | 748 | 147 | 110 | 374 | 18 | 129 | 132 |
| 4 | 4.9 | 965 | 189 | 142 | 482 | 24 | 166 | 176 |
| 5 | 6.0 | 1182 | 232 | 174 | 591 | 29 | 203 | 220 |
| 6 | 6.9 | 1359 | 267 | 200 | 679 | 33 | 233 | 264 |
| 7 | 7.8 | 1536 | 302 | 226 | 768 | 38 | 264 | 308 |
| 8 | 9.3 | 1831 | 360 | 270 | 916 | 45 | 315 | 351 |
The Uniform CFM column assumes a uniform air flow through the whole pipe, which is obviously incorrect. The Total CFM equal to the sum of the Annulus and the Inner zone, which comes out pretty close to the Rated values in the last column, taken from a comment by the seller.
Hard to believe I did the figuring before finding the “right” answers.
This is, admittedly, in free air without ducts or elbows, so the results will be lower when everything gets hooked up.
The Smell of Molten Projects in the Morning 