Ed Nisley's Blog: Shop notes, electronics, firmware, machinery, 3D printing, laser cuttery, and curiosities. Contents: 100% human thinking, 0% AI slop.
Autodesk just Borged Netfabb and, in the process of merging their address lists, asked me to update my info and agree to their very detailed Privacy Statement. You should take a look at it; the link will open in a new tab / window / whatever, so you don’t lose your place here.
Have you noticed how those “statements” always have a very long and firmly fixed line width that doesn’t adapt to your window size, use various shades of light-gray-on-white typefaces in the smallest sizes, and continue for pages and pages. I don’t believe in coincidences, either.
Some web browsers may transmit “do-not-track” signals to websites with which the browser communicates. Because of differences in how web browsers incorporate and activate this feature, it is not always clear whether users intend for these signals to be transmitted, or whether they are even aware of them.
Participants in the leading Internet standards-setting organization that is addressing this issue are in the process of determining what, if anything, websites should do when they receive such signals. Autodesk currently does not take action in response to these signals. If and when a final standard is established and accepted, we will reassess how to respond to these signals.
For information about cookies, web beacons and similar technologies, please read our Cookie Policy.
After plowing through much of their “statement”, I decided Autodesk doesn’t do anything I need to know about and, seeing as how Netfabb gradually faded from my attention when their web service joined Microsoft’s Azure cloud, I declined to “confirm my preferences” and didn’t click the big blue button. I doubt such inaction will remove my email address from their list, but it’s the only choice they offer.
Because I can’t tell if a website really wants to track me, I block ads, disable Flash, and destroy all cookies when I leave their site, Just In Case they inadvertently deployed all that crap. I’m sure they never intend to serve malware through an ad slot brokered on their site, but mistakes do happen, and I’m glad to assist them.
If you’re seeing ads on this page right now, they come from WordPress and I get a small cut. You should start using an ad blocker right now; if your browser doesn’t permit you to block ads, change browsers. If you worry that reducing my advertising revenue will compromise the quality and quantity of what you see here, send me a sack of money through Paypal. Fair enough?
On 12 July 2015, I sent a report to NYSDOT about how the traffic signals at Burnett Blvd / Rt 55 greenlighted opposing traffic when our bicycles were still in the intersection:
Can you increase the minimum green and yellow times on the signals from Burnett Blvd to Rt 55?
The current settings are too short for bicycle traffic making a left turn across six traffic lanes.
The pictures show key points from our ride on 2015-07-10, returning from the Balloon Festival in Poughkeepsie. We took the DCRT around Poughkeepsie, went through Arlington to Rt 376 at Collegeview, then took Rt 376 Red Oaks Mill.
The image sequence numbers identify frames extracted from video files. The Front camera runs at 60 fps and the Rear camera at 30 fps.
The red signals are in the process of turning off in Front 0196.
Burnett at Rt 55 Signal – Front 0196
One second later in Front 0260, the car and our bikes are starting to roll. Given the number of drivers blowing through red signals at full speed, devoting one second to watching for oncoming traffic seems prudent.
Burnett at Rt 55 Signal – Front 0260
The yellow signals are turning on in Front 0633, seven seconds after the green. The car has reached the pedestrian ladder across Rt 55, but we’re still crossing the westbound lanes of traffic. We may not be the fastest riders on the road, but we’re not the slowest, either.
Burnett at Rt 55 Signal – Front 0633
We’ve reached the far side of the intersection in Front 1142, just under 16 seconds from the green.
Burnett at Rt 55 Signal – Front 1142
However, Rear 0408 shows that the opposing signals turned green while we’re still crossing the eastbound lanes of Rt 55. That’s about 15 seconds after the Burnett Blvd signals went green.
Burnett at Rt 55 Signal – Rear 0408
About 2.7 seconds later, Rear 0490 shows cars accelerating across the intersection toward us as we cross the pedestrian ladder. They started rolling immediately after their signal went green; waiting a second isn’t a universal practice.
Burnett at Rt 55 Signal – Rear 0490
Setting the minimum Burnett green to 12 seconds, the minimum yellow to 10 seconds, and the minimum delay from Burnett green to Rt 55 green to 30 seconds would help cyclists (just barely) reach the far side of the intersection before opposing traffic starts rolling.
Also: can you adjust the sensor amplifiers on Burnett to respond to bicycles and mark the coil locations on the pavement in both lanes? That would help us through the intersection during low-traffic-volume times, as our bikes seem unable to trip the signals.
Thanks…
This reply from the NYSDOT autoresponder was all I ever got from them:
Thank you for your inquiry. We will respond to your email message as soon as possible.
On 2 August 2015, I sent a report to NYSDOT about how the traffic signals at Old Post Rd – Spring Rd at Rt 9 greenlighted opposing traffic when our bicycles were still in the intersection:
The minimum green-to-opposing-green signal timing from Old Post Road across Rt 9 to Spring road is about 18 seconds: not long enough for bicycles to safely cross an intersection with eight traffic lanes.
The “Green” picture shows our starting position as the signal turned green: behind the first car in line. There’s another car behind us, which ensures the loop sensor will trip; it does not detect bicycles.
Spring Rd – Rt 9 – 2015-08-01 – Green
The “Yellow” picture shows the signal changing after 12 seconds, with the car from behind us now in the middle of the northbound lanes. We’re still in the middle of the southbound lanes.
Spring Rd – Rt 9 – 2015-08-01 – Yellow
The “Opposing Left Green” from the rear camera, 18 seconds from the first picture, shows green left-turn arrows for Spring Road. The opposing cars began rolling with Mary lined up with the northbound right-turn lane and me lined up with the right travel lane.
Spring Rd – Rt 9 – 2015-08-01 – Opposing Left Green
The car behind blew through the red signal on Old Post Rd; I think that’s why the opposing left-turning cars didn’t start sooner.
In the other direction, I often use the left turn from Spring Rd to southbound Rt 9 to reach the South Road Square strip mall. Similarly short yellow and overall cycle times apply in that direction.
Can you add (at least!) five seconds to the yellow and perhaps ten seconds to the minimum cycle time for both directions? That would help us clear the intersection before opposing traffic starts moving again.
Can you also mark the sensor loop locations in all those lanes so cyclists can find them and adjust the amplifier sensitivity / dwell to respond to bicycles? We’ve lined up atop the quadrupole loop pavement cuts on Old Post Road to no avail, but there’s not even a hint of the loop positions under the new Spring Rd paving.
Thanks…
This reply from the NYSDOT autoresponder was all I ever got from them:
Thank you for your inquiry. We will respond to your email message as soon as possible.
On 6 January 2016, this email message arrived from the same email address that never responded to my reports (emphasis added):
Dear Mr. Nisley:
This is in response to your correspondence regarding your experiences as a bicyclist at the intersections of Route 55 at Burnett Boulevard and Route 9 at Spring Road in the Town of Poughkeepsie, Dutchess County.
The New York State Department of Transportation (NYSDOT) is in the process of investigating alternate detection types and inductance loop patterns that would detect a wider range of vehicles. As alternate detection types are tested and approved, they will be integrated into the next traffic signal upgrade at both intersections. The distance varies based on geometry. The loops are centered in each lane and the front loop is a quadrapole, so there are wires down the middle of the loops.
A new timing program was implemented at Route 9 at Spring Road in August, and the yellow and red clearance times meet the current standards. The timing at Route 55 at Burnett Boulevard is in the process of being updated, and the clearance times will be updated as necessary to meet the current standards. Clearance times are determined based on speed, intersection dimensions, grade, and reaction time and cannot be adjusted. The sensitivity on all loops will also be adjusted, so they are as sensitive as possible without causing cross talk between the loops.
We appreciate and share your interest in making our highway systems safe and functional for all users.
If you have any questions or need additional information, please feel free to contact our Regional Traffic Safety & Mobility Group at (845) 437-3396.
NYSDOT Hudson Valley Region
I don’t regard that date a coincidence; NYSDOT was not responding to my reports. I sent a further note to clarify a few points:
On 01/05/2016 02:18 PM, dot.sm.r08.nysdot wrote: Clearance times are determined based on speed,intersection dimensions, grade, and reaction time and cannot be adjusted.
That seems to mean the times can be adjusted, but you won’t adjust them to allow cyclists enough time to clear the intersection.
We appreciate and share your interest in making our highway systems safe and functional for all users.
So, giving opposing traffic a green light while we’re still in the intersection NYSDOT’s way of “making our highway systems safe and functional for all users”.
Do I understand your statements correctly?
No reply, as I’ve come to expect by now.
I think the emphasis on “meet(ing) the current standards” is how NYSDOT will attempt to defend against claims that road conditions caused or contributed to a car-on-bike collision. I find it surprising that contemporary “standards” would allow greenlighting opposing traffic against bicycles, but perhaps they simply choose a standard that excludes bicycles.
Mary wonders if the designers scaled the grips and spring tensions to suit women’s hands. Her experience shows that “tools for men” are too big and require too much grip strength for her comfort; applying pink plastic won’t improve them in the least.
So Time Warner updated the infrastructure upstream of Mary’s folks and installed a new cable modem / router, which killed my remote access using ssh (with RSA keys, passphrases, nonstandard ports, fixed internal IP addresses, port forwarding, port triggers, and all the Right Stuff). I just spent a harried pair of days trying and failing to figure out how to make this work again.
My laptop can ssh into my file server from our house network, both wired and wireless. Ditto when it’s on the Squidwrench Operating Table. Ditto from the low-quality Hampton Inn WiFi near her folks. Plunked on their desk and jacked into their router, however, that outbound ssh times out somewhere between their bits and my basement.
I dinked with the TW Surfboard modem / router, added the appropriate port forwarding & triggers, dialed back the firewall intensity, and ssh flat out doesn’t work in either direction from any PC (all running various Linus flavors). No diagnostics, no logs, nothing that I could find.
From the outside (our house or the Hampton), there’s no response from the PCs inside (on their desk). I’m not trying a loopback from inside to inside, which I know doesn’t work with consumer-grade routers. I’d planned to ssh from there to my basement file server, then ssh back to verify that the connections worked, but the outbound connection doesn’t work.
Probably unrelated, but equally frustrating: trying to configure Thunderbird’s outbound SMTP with their email server flat-out doesn’t work. Either the username / password isn’t valid (it is), various combination of ports / security / encryption (including the ones in the TW FAQs) don’t survive the configuration test, or a seemingly valid configuration doesn’t actually transmit email. Incoming email works only in IMAP mode, not POP3.
I finally set up outbound TW email to bankshot through his Gmail account, which will probably have unforeseen side effects.
The usual Google searches were unavailing, other than several notes suggesting that if you have any other choice of ISP or email provider than TW, do that. But it’s not like they have any choice; Verizon provides 1 Mb/s (!) DSL in that area and satellite Internet isn’t going to happen in an apartment.
Obviously, I’m doing several things wrong, but I have no idea what else to try. I’ve set up email and remote access often enough to get a whole bunch of things right, but that sure didn’t help with TW.
It’s one of those 1920-ish things with the impeccable stonework and bronze casting that you couldn’t possibly duplicate nowadays. But, at least twice between then and now, somebody thought it’d be a Good Idea to decorate it with what look to be Genuine Christmas Tree Lights:
Provincetown Pilgrim Memorial – detail
The most recent lamps and wires seem to be restrained by plastic clips glued onto the face of the stone:
Provincetown Pilgrim Memorial – lamp detail
A previous generation drilled small holes and inserted metal pins that didn’t survive in a salt-spray environment, so I guess plastic seemed like the right answer.
Here’s what the 0.35 mm diameter nozzle of my Makergear M2 looks like when printing a 0.40×0.25 mm thread on borosilicate glass with a coating of hairspray:
M2 V4 nozzle – thinwall box first layer
The dimensions:
Extrusion Dimensions
Some common household objects at the same scale:
Objects vs Thread Comparison
The accuracy required is literally hair-fine: being off by the diameter of the hair on your head can wreck the first layer of the printed object.
One turn of the M3 screws supporting the M2 platform move the mounting point by twice the thread thickness. Their positions on the platform amplify the motion by about a factor of two, so if you’re tweaking the screws by more than 1/6 turn at a time, you’re overdoing it.
For first-layer nozzle-to-platform distance adjustment:
If it increases by 0.25 mm, the plastic won’t touch the platform
If it decreases by 0.25 mm, the plastic won’t come out of the nozzle
For platform alignment:
If your printer can’t maintain the proper gap to within ±0.10 mm across the entire platform, it won’t produce accurate results
Works for me, anyhow. All I do is slice whatever object I’ve just designed, turn the M2 on, and print it. No muss, no fuss, no wasted motion: It Just Works.
The sketches come from my Digital Machinist column (DM 10.4). They’ve been covering a bunch of 3D printing topics, so if you’re interested in that kind of stuff…
If you’re running Windows, then you have more experience than I do, but it seems Microsoft, for reasons best known to it, really really really wants you to upgrade to Windows 10, has been forcing nagware onto every Windows box in existence, and actively working to defeat efforts to remove said nagware.
Our Token Windows Box, an off-lease Dell Optiplex 780 that arrived bearing Windows 7 Professional, will never, ever get upgraded, because it’s running a bunch of ancient Windows programs that interface with specific bits of hardware, none of which (most likely) will work with Windows 10. In any event, I see no reason to go through the hassle of “upgrading” an old machine, (maybe) resolving all the inevitable compatibility problems, and (maybe) having no way to roll back the upgrade, all for a few programs run, at most, monthly.
Continually declining Windows 10 upgrade prompts isn’t my idea of a Good User Experience, but I’m also tired of manually inspecting and killing updates that re-re-re-install the nagware.
The GWX Control Panel (“GWX” = “Get Windows 10” in MS-speak) seems to be the least awful way of dealing with this mess. It’s not offered by Microsoft, for obvious reasons, but is offered free-of-charge.
The Smell of Molten Projects in the Morning 