#linuxcnc Logs
Oct 24 2022
#linuxcnc Calendar
01:51 AM Deejay: moin
02:52 AM mrec: did anyone ever use sc32 / sc40 chinese pneumatic cylinders?
03:55 AM travis_farmer[m]: G'Morning
04:40 AM Tom_L: morning
05:08 AM JT-Cave: morning
05:11 AM pere: uhi
05:11 AM pere: hi
07:36 AM solarwind: mrec I have
07:43 AM JT-Cave: lol aliexpress sc40x1000 $20 shipping $80
07:44 AM solarwind: Yeah it's not cost effective anymore
07:44 AM solarwind: I remember the days when the shipping was free for most things
07:52 AM travis_farmer[m]: used to be worth the money to order electronic parts in bulk from china... if you could handle the wait. but all too often, i have seen free shipping no longer an option.
07:54 AM mrec: solarwind: which tube do you use? 8mm / 10mm / 12mm?
07:54 AM solarwind: mrec you mean cylinder diameter?
07:54 AM mrec: no the tubing for the cylinder
07:55 AM mrec: I think SC40 should have 1/4"
07:55 AM solarwind: oh the pneumatic hose?
07:55 AM mrec: currently I settled for an 8mm tube
07:55 AM mrec: yes
07:55 AM solarwind: That entirely depends on your needs
07:55 AM solarwind: it can be anything
07:55 AM solarwind: I use mine as hold down clamps, so velocity is not a factor
07:55 AM mrec: I wonder if there's any disadvantage bigger vs smaller tubes
07:55 AM solarwind: Yes
07:56 AM solarwind: if you need high speed actuation, use larger diameter hoses
07:56 AM solarwind: if you don't need that kind of air flow, then it doesn't matter
07:56 AM solarwind: I used what I have, which is 6mm
07:57 AM mrec: ok, I just want to change discs automatically on my milling machine for cutting gears
07:57 AM solarwind: For hold down clamps, that's plenty, since that application only need pressure
07:57 AM mrec: I'm about to buy SC32 and SC40-25mm
07:57 AM solarwind: well if you want faster actuation, increase pressure and flow
07:58 AM solarwind: so give it a nice fat 12mm hose and crank up the pressure to the max permitted by the cylinder
07:58 AM mrec: I'll put the tailstock on linear rails on the milling machine
07:58 AM mrec: and squeeze the part into the A axis
07:58 AM mrec: I'm getting tired after making 400 gears
07:59 AM solarwind: when the cylinder actuates from one end to the other, keep in mind that it has to evacuate the air from the other side too
07:59 AM solarwind: and that air isn't at atmospheric pressure
07:59 AM solarwind: so make sure your valves are big enough too
08:00 AM solarwind: that actuation speed is entirely proportional to hose diameter and valve air flow restrictions
08:00 AM mrec: aliexpress.com/item/1005003777549269.html
08:00 AM mrec: I'll use this for playing
08:01 AM solarwind: Yeah that's fine for testing, but that's certainly going to be the main restriction for anything above 8mm hoses
08:01 AM solarwind: I know because I have one
08:01 AM mrec: aliexpress.com/item/1005003539781577.html and a similar one for controllig
08:01 AM mrec: controlling
08:02 AM solarwind: don't let those screw threads deceive you. The actual air inlet/outlet diameter is not as big as the pipe thread on those
08:03 AM solarwind: Not sure how fast you want to go, but pneumatic cylinders are very fast by nature to begin with, so I'm pretty sure all this is a non-issue
08:03 AM mrec: I have the electrical one here already (I used it for other things in the past) that one is fine
08:03 AM mrec: I just never used a pneumatic cylinder before especially not for clamping anything
08:04 AM solarwind: well it's nothing like a hydraulic cylinder lol
08:04 AM solarwind: it'll snap from one end to the other in milliseconds
08:04 AM ZincBoy[CAON][m]: Usually the trick is slowing them down so they don't slam into limits of travel.
08:04 AM solarwind: yes
08:04 AM solarwind: let's just say even the smallest one will knock you out if you point it at your head
08:04 AM solarwind: don't ask me how I know
08:05 AM ZincBoy[CAON][m]: I use adjustable flow control valves on both ports to control the actuation speed.
08:05 AM solarwind: yup same
08:05 AM mrec: what was that movie called... no country for old man?
08:05 AM solarwind: yes
08:05 AM mrec: reminds me about that one
08:05 AM solarwind: that's what came tom ind too
08:05 AM ZincBoy[CAON][m]: Don't point it at your head 😛
08:05 AM solarwind: but they use the same thing for knocking out cattle
08:06 AM solarwind: I have a 5 meter long pneumatic cylinder sitting here
08:06 AM solarwind: I plan to use it as the main actuator for my garage door
08:07 AM solarwind: Someone just gave it to me
08:09 AM mrec: does anyone automatically check the quality of parts during manufacturing?
08:09 AM solarwind: oh ho ho ho nooooooo
08:09 AM ZincBoy[CAON][m]: I use valves like this to control the cylinder actuation rate: https://www.festo.com/ca/en/p/one-way-flow-control-valve-id_GRLA_QS/?q=~:festoSortOrderScored
08:09 AM solarwind: not in China
08:09 AM JT-Cave: Tom_L, https://github.com/jethornton/mesact/raw/ini4/mesact_0.7.4_amd64.deb
08:09 AM JT-Cave: ready for testing
08:09 AM solarwind: There's is a guy with a roll of "QC passed" stickers though and his job is to stick them on everything
08:10 AM mrec: no I mean if you do it yourself
08:10 AM mrec: I do some manufacturing myself (my own parts) ... and I see all the tired work
08:11 AM ZincBoy[CAON][m]: How tall is your door? A 5m cylinder is a big beast! Need a big air receiver to actuate that.
08:11 AM mrec: I'm measuring parts one by one and line them up so I see if the quality is getting worse
08:11 AM solarwind: I recommend you add a roll of those "QC passed" stickers to your aliexpress order
08:11 AM mrec: ZincBoy[CAON][m]: a castle ^^
08:11 AM ZincBoy[CAON][m]: @mrec, yes. SPC and other systems are in use. China will if you pay for it and keep watch.
08:11 AM solarwind: ZincBoy[CAON][m] I have a 1,000L air tank
08:11 AM solarwind: and a 11kW compressor
08:12 AM solarwind: you should come by my hobby workshop at some point if you're in the Scarborough area
08:12 AM mrec: I don't want to manufacture my things in China I do them myself
08:12 AM solarwind: mrec I know I was just making a joke
08:13 AM ZincBoy[CAON][m]: I do in process probing on my machine and automatically adjust offsets to keep parts in spec. With that I can hold +/-0.0005" on critical dimensions without any issues.
08:13 AM solarwind: wait no, it's 1,600L
08:15 AM solarwind: its volume is far greater than the cylinder, and I keep it at around 13bar pressure
08:15 AM solarwind: really helps when you want to clean out your car with a very high flow air blow gun
08:16 AM mrec: I'm just curious does anyone here manufacture in China?
08:16 AM mrec: the quotes which I got in the past were quite expensive .. including shipping it wasn't really worth it
08:16 AM mrec: lasercutting was okay so far
08:16 AM mrec: some sheet metal, but bigger parts no
08:18 AM mrec: I know from accounting that some US companies order their injection molding tools in Taiwan. That seems to be worth it.
08:34 AM JT-Shop: Tom_L, mcmaster sells cdi torque wrenches
08:52 AM pere: roguish[m]: any news on the div2 component?
09:32 AM Tom_L: JT-Shop, what parts need testing most? load and save ini?
09:32 AM Tom_L: mine goes from 10-100 ftlb
09:32 AM Tom_L: probably fairly standard range
09:40 AM JT-Shop: Tom_L, yes load and save then change axis letters, add/remove axis, add/remove spindle sscard etc
09:43 AM JT-Shop: create your own sections add key/value pairs and comments to a section
09:45 AM JT-Shop: everything from the start of a section to the start of the next section is part of the section so if you put a comment before a section it belongs to the previous section
09:45 AM JT-Shop: a blank line before a section is recommended
09:53 AM Tom_L: you've been busy
09:54 AM JT-Shop: very busy
09:55 AM JT-Shop: the joint axis section was interesting
09:56 AM Tom_L: i wonder if you should keep the sserial.hal but just remove reference to it in the other files
09:57 AM Tom_L: probably not that common but i did testing with one and kept the setup files for it for a while in case i wanted to go back and review it
09:58 AM Tom_L: i think it's fine either way
09:58 AM Tom_L: most won't be doing that
10:03 AM Tom_L: ok, upon loading it says Smart Serial file Loaded but there isn't one. i removed it with the tool
10:03 AM Tom_L: and the filelist.hal doesn't show it
10:04 AM Tom_L: created a 7i84 one then removed it
10:05 AM JT-Cave: filelist.hal is only for HALFILE but I see I forgot that in ini4
10:07 AM Tom_L: i did that without restarting mesact...
10:07 AM Tom_L: just reloaded the ini
10:07 AM JT-Cave: hmm looks like I need to fix buildini.py it's still the old way with multiple HALFILE entries
10:07 AM Tom_L: custom section looks ok
10:08 AM Tom_L: also a reminder about multiple MDI_COMMAND entries... i'm sure you've been to busy to address that yet
10:09 AM JT-Cave: those would have to be in a separate hal file and called by filelist.hal
10:10 AM JT-Cave: I think...
10:11 AM Tom_L: i did test it too and it has to be MDI_COMMAND = not MDI =
10:13 AM Tom_L: fixing MDI_COMMAND isn't a priority
10:13 AM Tom_L: most noobs wouldn't know what that was
10:14 AM JT-Cave: hmm I need to fix the hal section of the ini in master then cherry pick that to ini4
10:15 AM Tom_L: yeah, that's above my paygrade :)
10:16 AM JT-Cave: just pushed to master the fix for MDI_COMMAND
10:16 AM Tom_L: stupid school... kid had an OSHA10 class to take to finish his 2 degrees so he signed up for it and nobody notified him. they called today at his work and wondered why he wasn't in class
10:17 AM Tom_L: just started his new job a few weeks ago..
10:17 AM Tom_L: luckily they were understanding and let him go
10:19 AM JT-Cave: always something
10:20 AM JT-Cave: time to go to the sawdust factory
10:20 AM perry_j1987: morning
11:48 AM * roycroft waves goodbye to all the other networks
12:34 PM CaptHindsight[m]: https://www.grobinc.com/cold-rolled-products/standard-spline-shafts/ doesn't matter which size you want, they might have 0-150ft in stock with the balance in 16-20 weeks
12:39 PM CaptHindsight[m]: and they don't seem to understand specifications like having a shaft 5ft long, since they roll them in 6ft lengths they wanted to know if that would work. It could work if you make the machine 1ft longer or have the 1 ft of extra shaft stick out from the front of the machine.
12:39 PM CloudEvil: Loctite red - meh, round is fine too.
12:39 PM CloudEvil: (alas not quite actually)
12:45 PM CaptHindsight[m]: https://www.surpluscenter.com/Power-Transmission/Shafting/ their hex shaft is priced way higher than ordering it from a mill
12:46 PM CloudEvil: Will the mill actually sell you ones though
12:48 PM CaptHindsight[m]: sure, they make money by selling stuff
01:00 PM roycroft: make money selling stuff?
01:00 PM roycroft: what a concept
01:00 PM roycroft: does that actually work?
01:01 PM CloudEvil: Often, no.
01:01 PM * CloudEvil looks at the abundent british steelmaking industry.
01:02 PM roycroft: even if you do it on the internet and want to make the money fast?
01:10 PM CaptHindsight[m]: casino vs stock market
01:37 PM Loetmichel: CloudEvil: wouldnt that be "past perfect" for any british industry?
02:09 PM solarwind: pcw-home thanks for making these I/O cards for non-obscene prices
02:10 PM solarwind: pcw-home do you approximately what the analog bandwidth on the 7I36 is?
02:27 PM pcw-home: About 2 KHz (sufficient for velocity mode analog servos)
02:28 PM solarwind: pcw-home thanks, what about for torque mode, have anything for that?
02:28 PM solarwind: ZincBoy[CAON][m] was using something for torque control at 4kHz, I forgot which card it was
02:31 PM pcw-home: No except maybe a 7I65 (torque mode also often needs a 4 KHz or so servo thread which may be difficult with some hosts)
02:32 PM solarwind: won't be an issue with my system
02:32 PM solarwind: I can do > 16kHz
02:32 PM solarwind: I don't necessarily intend to run it with LinuxCNC
02:45 PM ZincBoy[CAON][m]: @solarwind, that was using an 8i20 BLDC drive board and the control is all over sserial.
02:49 PM solarwind: Oh damn, I didn't realize that even existed
02:49 PM solarwind: that might be an option instead of other servo drives
02:50 PM solarwind: the site could use some work lol
02:50 PM solarwind: I can't find any references to the 8C20
02:50 PM solarwind: That is, can't find the 8C20 page itself
02:51 PM ZincBoy[CAON][m]: No, I couldn't find anything on the 8C20 other than the mention in the 8i20.
02:51 PM pere: hi
02:51 PM solarwind: ZincBoy[CAON][m] what is driving the 8i20 in your use case?
02:52 PM ZincBoy[CAON][m]: I was using it for the spindle drive in my NM145 mill. Replaced the original ultra crap drive and got full servo control over the spindle.
02:52 PM solarwind: I meant what is feeding the 8i20 with serial data
02:53 PM solarwind: is it the FPGA card relaying information back to the host which is doing the velocity PID loop?
02:53 PM ZincBoy[CAON][m]: It was one of the PCIe Mesa cards. The host is doing all of the control loop, the fpga is just handling the sserial comm.
02:54 PM solarwind: Ok yeah that makes sense. That would work perfectly for me
02:54 PM Rab: 8C20 references on the forum: https://forum.linuxcnc.org/27-driver-boards/35379-mesa-8c20 https://www.forum.linuxcnc.org/27-driver-boards/34048-what-is-the-secret-card-8c20
02:54 PM solarwind: Apart from it being out of stock, I wish the documentation was more obvious
02:54 PM solarwind: Because the price is right, I already have the motors
02:55 PM solarwind: And I can do the serial control from a dedicated microcontroller if needed
02:56 PM solarwind: STM32F7 with FreeRTOS would easily handle that PID loop
02:57 PM pere: looking for a controler example in the configurator, it occured to me that it would be useful with a free text search option to search for hb04. :)
02:59 PM solarwind: ZincBoy[CAON][m] lol response from DMM: Note that our EtherCAT servo drives are not in production yet so it's not available for sale. The proprietary IC needed for EtherCAT is heavily delayed production and we cannot purchase these chips to make the drives. The EtherCAT development is finished, we just cannot manufacture the drives yet.
02:59 PM solarwind: what the hell, proprietary IC?
02:59 PM solarwind: EtherCAT is supposed to be open
03:00 PM ZincBoy[CAON][m]: Just means the specific ASIC they used is not available. Typical these days with the semiconductor supply chain in a mess.
03:00 PM solarwind: yeah
03:01 PM W1N9Zr0[m] is now known as W1N9Zr03887[m]
03:02 PM CaptHindsight[m]: open as in not open
03:03 PM ZincBoy[CAON][m]: Open just means the spec is open. It doesn't mean there is non-proprietary hardware available. Or that the hardware itself is open.
03:03 PM solarwind: Yeah I know, I was just more commenting on why the hell they're using proprietary crap
03:04 PM ZincBoy[CAON][m]: I wouldn't want to roll my own low latency ethernet phy if there was an off the shelf solution.
03:05 PM solarwind: the Ethernet PHY isn't special
03:05 PM solarwind: these days they're all very low latency
03:05 PM solarwind: it's the protocol stack and DMA setup
03:06 PM solarwind: and why they didn't standardize on 100BASE-"optical" is beyond me
03:06 PM solarwind: it's so much better than BASE-T in every way
03:06 PM ZincBoy[CAON][m]: Yes, and if there is an asic that bundles it all together then that is how I would go if I were them.
03:06 PM ZincBoy[CAON][m]: Optical has its own issues. Mostly with fiber termination and connector care.
03:07 PM solarwind: Yeah, never had an issue in practice with multimode fibre
03:07 PM solarwind: can DIY terminate and rub it on your shirt to clean it
03:07 PM solarwind: or just buy the patch cable for $2
03:08 PM roycroft: i've hardly ever used multimode fiber
03:08 PM ZincBoy[CAON][m]: LOL. Don't try that at home kids. The company I work for makes optical transceiver ICs. If you want a flakey link that is how you get it.
03:08 PM roycroft: single mode is a bit more particular than that, i can attest
03:08 PM solarwind: Yeah single mode is far more sensitive
03:09 PM solarwind: I run 100GBASE-LR over single mode
03:09 PM roycroft: and long-haul fiber spans have their own issues, especially with a muxed signal
03:09 PM roycroft: since the different frequencies travel at different speeds
03:09 PM ZincBoy[CAON][m]: Multimode is still a 50-80um active area vs 8um for single mode.
03:09 PM roycroft: over the course of 50km or so the light gets "smeared"
03:09 PM ZincBoy[CAON][m]: That is why you stick some negative dispersion fiber in there 🙂
03:09 PM solarwind: all of this is a non issue for the 50cm patch cable you need to link servo drives
03:10 PM roycroft: yes
03:10 PM roycroft: which is the most expensive component of a fiber node
03:10 PM roycroft: and it's a completely passive device
03:10 PM solarwind: It's _really_ hard to get a flaky signal for short spans
03:10 PM roycroft: and yes, for a 50cm cable that is completely irrelevant
03:10 PM roycroft: but it's still interesting
03:10 PM ZincBoy[CAON][m]: If you are only going 50cm then you really don't need optical.
03:11 PM roycroft: and even if you do want to go optical, you don't need fiber
03:11 PM ZincBoy[CAON][m]: Most datacenter are still direct attach copper below 5m.
03:11 PM roycroft: you can just point a laser at it and shoot the sigal through the air :)
03:11 PM CaptHindsight[m]: who makes all the cheap VCSEL's in the Chinese fiber transceivers?
03:11 PM solarwind: for noise immunity it's nice
03:12 PM ZincBoy[CAON][m]: There are a few vendors for VCSELs but the low end is pretty variable.
03:13 PM CaptHindsight[m]: they were >$100 for years and then suddenly they got low cost
03:13 PM solarwind: I was kidding about the rub it on your shirt comment by the way. It'll work, but I still have a specific tool to clean them. It's one of those push pen type things
03:13 PM ZincBoy[CAON][m]: With proper grounding and attention to signal routing a differential electrical signal is going to be just fine. If you have enough EMI/ground issue that you are killing a 50cm electrical link, fibre won't help you.
03:15 PM solarwind: Yeah that is true
03:16 PM ZincBoy[CAON][m]: @CaptHindsight, what datarate? I know that most of the 10G modules went from active retiming to just a laserdriver/TIA. This brought the cost down quite a bit. Plus the high end uses higher bandwidth VCSELs and PDs. So the price crashed on the low bandwidth units.
03:16 PM solarwind: But wave one of those arc lighters near a well shielded cable and watch the packet loss counters
03:16 PM CaptHindsight[m]: 1Gb
03:16 PM solarwind: It's fun
03:16 PM ZincBoy[CAON][m]: The same thing is happening in the 25G NRZ space.
03:18 PM ZincBoy[CAON][m]: Oh, 1G I haven't looked at for years. Those will be hitting the min value for the physical SFP module casing and components.
03:18 PM solarwind: I'm buying 100GBASE-LR transceivers on ebay for $6 each
03:19 PM solarwind: they go nicely with my $600 36 port QSFP28 switch
03:19 PM ZincBoy[CAON][m]: @solarwind, yes because those arc lighters would never be allowed be sold if they had to pass regulatory approval. They are effectively a spark gap transmitter.
03:20 PM solarwind: similar thing from motor brushes
03:20 PM ZincBoy[CAON][m]: Yes, you can get stuff off ebay for cheap but look at the new prices. A new 100g module is still in the $50-100 range.
03:20 PM solarwind: not nearly as bad but still
03:20 PM solarwind: 100GBASE-SR yes, but LR4 is still $400
03:22 PM solarwind: https://www.fs.com/products/102531.html?attribute=12760&id=394766
03:22 PM ZincBoy[CAON][m]: If your shielded cable is getting errors from a brushed motor there is something wrong. Unless you are winding the cable around the motor that should not cause issues. I run a stick welder and HF-TIG torch within 20ft of my mill and it doesn't even hiccup. No packet loss on the UTP ethernet in the shop either.
03:23 PM CaptHindsight[m]: are they making any multimode 1Gb transceivers that can be used over more than a few meters?
03:23 PM solarwind: CaptHindsight[m] few metres?
03:23 PM solarwind: You mean few thousand metres?
03:23 PM CaptHindsight[m]: or did they 10 -15 years ago?
03:23 PM solarwind: I have a bag of them with like 500 transceivers
03:24 PM CaptHindsight[m]: low cost multimode used to have a very short range, plastic fiber
03:24 PM CaptHindsight[m]: lost cost = few $ per transceiver
03:25 PM CaptHindsight[m]: I didn't follow the evolution from the late 90's untilnow
03:25 PM CaptHindsight[m]: now 10GB is cheap and has a long range
03:26 PM CaptHindsight[m]: even fiber on your PC sound cards were kept not low priced
03:26 PM ZincBoy[CAON][m]: OM3 and OM4 multimode are good for at least 1km at 1g
03:27 PM ZincBoy[CAON][m]: Those are glass fibers though and not the plastic TOSLINK type.
03:28 PM CaptHindsight[m]: S/PDIF with plastic fiber
03:30 PM CaptHindsight[m]: I'll read some fiber history for the past 20 years
03:31 PM CaptHindsight[m]: solarwind: this was back when x86 did not reach 1GHz core speeds yet
03:31 PM CaptHindsight[m]: the fastest area of the PCB was the fiber link
03:32 PM roycroft: but was it back when nobody would "ever" need more than 640k of ram?
03:36 PM CaptHindsight[m]: nah by then we had loads of RAM, 32MB or more :)
03:36 PM CaptHindsight[m]: and 10GB hard drives
03:37 PM solarwind: I know, I remember
03:37 PM pere: heh, I just noticed, /lib/udev/rules.d/99-xhc-whb04b-6.rules seem a bit over the top. It provide access to every user on the computer (mode 666), all members of the plugdev group and everyone logged in via the console (uaccess). While only the latter is recommended, any of these access methods would be sufficient...
03:37 PM solarwind: I had, and still have that 5GB quantum bigfoot drive
03:44 PM Rab: ahahaha Quantum Bigfoot
03:45 PM Rab: I was working in a computer shop when those came out...impressive RMA rate
03:48 PM Rab: We theorized that the Bigfoot was a strategy for Quantum to dump their 5.25" platters for cheap while the market was going completely to 3.5".
03:58 PM roycroft: back in the 80s we had a huge percentage of seagate hard drives fail prematurely
03:58 PM roycroft: iirc they were the st-251
03:58 PM roycroft: we finally discovered that one of the mounting screw holes on the side of the drive was not planar with the other three
03:59 PM roycroft: and that if we mounted the drive with three screws it would be fine
03:59 PM roycroft: but with all four they would fail almost immediately
03:59 PM CloudEvil: Die permenantly?
03:59 PM roycroft: the heads would crash
03:59 PM CloudEvil: ah
04:00 PM roycroft: which makes sense, since installing that fourth screw would twist the drive
04:01 PM roycroft: no, it wasn't the st-251
04:01 PM roycroft: it was the st-238
04:02 PM roycroft: 32MB mfm drive
04:02 PM roycroft: 5-1/4" half height
04:02 PM roycroft: and you can still buy that drive
04:03 PM roycroft: https://www.amazon.com//dp/B00DH3BYY8
04:04 PM CloudEvil: fun
04:08 PM JT-Shop2: dammit the brake lines are in the way to get the differential out...
04:10 PM roycroft: chop them off
04:10 PM roycroft: brakes are overrated
04:11 PM roycroft: if you ever need to stop you can just go all thelma and louise
04:11 PM roycroft: you'll come to a stop eventually with gravity assist
04:18 PM JT-Shop2: I took the mount loose by the flex hose and I "think" I can wiggle it by...
04:19 PM * JT[m] uploaded an image: (212KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/zjkUYzfRUgoYVmGLzIRVumkC/20221024_161749_362235283495828828.jpg >
04:43 PM * JT[m] uploaded an image: (214KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/ebWEEMfZqUDelnTfhDYstBhR/20221024_164153_5934244733879788472.jpg >
04:58 PM Tom_L: easier to work on if you turned the whole thing upside down...
05:00 PM roycroft: folks used to make rotisseries for vw buses to restore the underside
05:00 PM roycroft: i'm sure that would be easy to do for a small car
05:02 PM JT-Shop2: actually nothing has been difficult about the job so far
05:04 PM JT-Shop2: a lift "might" have made it easier to remove some things but down near the floor made it easy to remove the bigger parts
05:05 PM * JT[m] uploaded an image: (239KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/BqXJKJXZcQPPqkMdvxCoSrQh/20221024_170350_5934064354454412875.jpg >
05:05 PM JT-Shop2: I'll wash it when it's back together...
05:06 PM Tom_L: not much left under there
05:07 PM JT-Shop2: 8 bolts to the prize
05:07 PM JT-Shop2: ordered a 5-75 foot pound from mcmaster
05:08 PM JT-Shop2: did you see they sell snap on?
05:08 PM Tom_L: should be good unless you start doing head work
05:08 PM Tom_L: no
05:08 PM Tom_L: i didn't look
05:08 PM JT-Shop2: <JT-Shop> Tom_L, mcmaster sells cdi torque wrenches
05:09 PM Tom_L: yeah i saw that
05:09 PM JT-Shop2: they emailed me with the model number and who makes it, they are good about doing that if you ask
05:09 PM Tom_L: i try to not max mine out but keep somewhere in the middle
05:09 PM Tom_L: years back the shops i was at had big ones
05:10 PM Tom_L: if it required something more that what i had
05:10 PM JT-Shop2: most of the torques are 20-50 fp on this job
05:10 PM Tom_L: not much goes beyond that really
05:10 PM JT-Shop2: on the rigs we had a 10-1 torque multiplier for the head nuts
05:10 PM Tom_L: maybe 75-80
05:11 PM JT-Shop2: 750 foot pounds each
05:11 PM Tom_L: yeah
05:12 PM JT-Shop2: I need to drop the transmission down another 1/4" I think to put the differential back on
05:19 PM JT-Shop2: I'm so glad to be at this point in the job
05:23 PM JT-Shop2: time to wander inside or to the catio
05:55 PM solarwind: So I have a small stepper motor on my desk with a machined aluminum timing belt pulley on it. I just like to fidget with it while I think
05:56 PM solarwind: Every time I do that and turn the pulley with my hands, my fingers get coated with this black stuff
05:56 PM solarwind: It's almost like I've been handling graphite rods
05:56 PM solarwind: The stepper obviously takes a bit of torque to turn, even though it's unloaded, because of reluctance and all that
05:57 PM solarwind: So there's some friction between my fingers and the pulley
05:57 PM solarwind: So is that the aluminum oxide formed on the pulley just constantly getting rubbed off and sticking to my fingers?
05:58 PM solarwind: The aluminum pulley is bright and clean. There's no coating or paint on it. It's just straight up machined aluminum. Probably 6061 or 7075
06:00 PM solarwind: Unleaded brass and steel never seem to do that to my hands. Leaded brass does though
06:50 PM roycroft: aluminium does that all the time - it can be a messy metal to work with
06:53 PM xxcodery: techinically all alum has a coating on it
06:54 PM xxcodery: its fairly rare to actually touch bare alum as it oxides over in minutes
06:57 PM roycroft: if you surface the aluminium in an oxygen-free environment it will remain uncoated until it is exposed to oxygen
06:57 PM roycroft: at least it won't be forming an oxide layer
06:57 PM roycroft: it may react with other elements in the oxygen-free environment
07:00 PM xxcodery: yeah. man is it reactive
07:00 PM xxcodery: if its oxide wasnt stable it would be very useless metal
07:00 PM roycroft: not as reactive as sodium
07:00 PM xxcodery: yep lol
07:00 PM roycroft: sodium just can't stand being alone
07:00 PM roycroft: it requires company or it throws big fits
07:01 PM xxcodery: lol
07:10 PM xxcodery: salts column is all like that
07:20 PM roycroft: so i just got something totally awesome in the mail
07:21 PM roycroft: i got a razor blade scraper on a long handle
07:21 PM roycroft: that in and of itself is nice, but not awesome
07:21 PM roycroft: what is awesome is that i got with it a box of 100 plastic "razor blades"
07:23 PM roycroft: they are the same form factor as standard safety razor blades, so they will fit in most devices that hold those blades
08:17 PM Rab: roycroft, I went out of my way to buy a pack of those plastic razor blades, but every time I try to use one I just end up finishing the scraping with my thumbnail. I'd say they're more of a booby prize.\
08:18 PM xxcodery: can always 3d print em too I guess
08:33 PM roycroft: i'll have a better sense of how brilliant they are shortly
08:34 PM roycroft: the plastic is pretty hard, and the edges pretty crisp
08:34 PM roycroft: i have high hopes
08:34 PM Tom_L: plastic...
08:37 PM Tom_L: rained a bit finally but not near enough
08:40 PM Tom_L: ceramic maybe...
09:18 PM solarwind: pcw-home is everything out of stock?
09:18 PM solarwind: Any ETA on something like HD-DB ADAPT? I can't imagine the chip shortage would affect a passive device like that
09:19 PM solarwind: Someone gave me a 5I25 but none of my decade old computers have PCI slots, so I need to use the 7I77 with the PCIe card I have
09:20 PM solarwind: pcw-home or is there a schematic available?
09:25 PM pcw-home: You can use a 50 pin FPGA card with a 7I77 with the correct firmware, basically just connect I/O 0..16 of the 50 pin card to I/O 0..16 of the 7I77
10:20 PM solarwind: pcw-home ah ok that's helpful, thanks. But what if I wanted to use the other 2 50 pin headers on the 6i24-25 as well?
10:20 PM solarwind: I'm guessing the 50 pin header is just a superset of the smaller header so it should just work right?
10:26 PM solarwind: But I guess I'll wait until you guys get some stock in. I like the specs on the servo amplifier
11:53 PM Rowow[m] is now known as Rowow5457[m]