[cctalk] Re: Philips P3800 key lock swtch
Maybe Toool would be a good place to ask https://toool.nl/Toool They do run a LockCon but it appears to be in October and costs around 360 Euro plus accommodation. So, an expensive route to find a key. The UK organisation run a monthly meeting at FizzPop in Birmingham which is more accessible but perhaps less likely to recognise the keyway. On Mon, Apr 7, 2025 at 9:42 PM Paul Koning via cctalk wrote: > > > > On Apr 7, 2025, at 2:34 PM, Dennis Boone via cctalk < > cctalk@classiccmp.org> wrote: > > > >>> Can you post a picture of what you can see? > > > >> Can't post photos here, > > > > Tony and I assembled a page of photos, which can be found here: > > > > https://yagi.h-net.org/philips_p3800_lock/ > > It looks like a Lips lock, which would make sense for a Dutch computer. > > paul
[cctalk] Re: Philips P3800 key lock swtch
On Tue, Apr 8, 2025 at 5:33 PM Paul Koning wrote: > > > > > On Apr 8, 2025, at 12:06 PM, Tony Duell via cctalk > > wrote: > > > > ... > > I think there is only one key. The lock is somewhat unsual though. > > > > There are 4 pins in the plug (the rotating part). There are 3 > > positions of the lock and the key is removeable in all 3. There are > > therefore 3 sets of upper pins and springs in the housing. The 3 > > positions are 'OFF' (anticlockwise), ON (centre), PM (clockwise). The > > visible holes in the housing were always empty, I've not removed > > anything there. I think they were there to allow the springs and upper > > pins to be inserted when the lock to be assembled. Note that the hole > > for the 'OFF' position are ofset away from the front of the lock. > > > > Here's what I think happens. In the ON position you can use the key > > and simply turn to PM. And go back again But to select OFF you have > > to push the key and plug inwards before turning the key. To turn on, > > you just turn the key (you can't go too far and select PM by > > accident), the plug will then spring forwards. There was a compression > > spring in the back of the lock, forcing the plug forwards. I have > > removed this now for the moment. > > > > The in/out motion of the plug is controlled by a pin (the end of which > > is the silver circle in the 'side' photo) running in a groove in the > > plug. > > That all makes sense. To allow removal of the key there have to be holes at > that position to let the pins (the ones in the cylinder) move upward as the > key is extracted. Then normally the pins in the lock body would be pushed > inward by the springs, locking the cylinder. > > If for some of the positions you have holes but they are empty, it means the > key can be removed in that position, and then the pins would probably drop > back into the cylinder, leaving it unlocked and free to turn. But in > addition, you could turn the lock upside down and let the pins slide out (be > sure to keep them in order). The lengths of those pins would tell you the > cuts needed for the key. I think all the holes have pins. The confusion arises because the lock is effectively upside-down. As viewed on the machine, the pins are at the lower edge of the keyway, The cut edge of the key goes in on the bottom. The holes on the 'top' are there for assembly. However you do raise an interesting idea. My intention was to knock out the pin that guides the moton of the plug from the other side, or if that's impossible to carefully drill it out. I was then going to pick or shim the lock to free it and remove the plug by pressing it from the rear with a suitable cylindrical follower to keep the 3 sets of upper pins and their springs in place. Then remove the pins from the plug and measure them in order to cut a key. But I could, instead, after picking or shiming the lock simply turn the plug through 180 degrees to line up the pins with one set of holes. And extract them that way. -tony
[cctalk] Re: Why I am not worried about AI taking over the world!
On Wed, 2 Apr 2025 at 17:54, Bill Gunshannon via cctalk wrote: > the COCO which > was never called a TRS-80 that I know of. Yes it was. Note the title of this page: https://en.wikipedia.org/wiki/TRS-80_Color_Computer Here's a pic: https://en.wikipedia.org/wiki/TRS-80_Color_Computer#/media/File:TRS-80_Color_Computer_1.jpg -- Liam Proven ~ Profile: https://about.me/liamproven Email: lpro...@cix.co.uk ~ gMail/gTalk/FB: lpro...@gmail.com Twitter/LinkedIn: lproven ~ Skype: liamproven IoM: (+44) 7624 227612: UK: (+44) 7939-087884 Czech [+ WhatsApp/Telegram/Signal]: (+420) 702-829-053
[cctalk] Re: Philips P3800 key lock swtch
Hi, > On 8 Apr 2025, at 09:17, Adrian Godwin via cctalk > wrote: > Maybe Toool would be a good place to ask I've already sent a mail off-list. I am the former president of Toool and organiser of LockCon for many years. And co-writer of the book about Locksport :) I can create a key, if need be a 3D printed one (but the profile looks rather standard in my neck of the woods). But I'd need to (temporarily) have the lock. The lock looks interesting because if seems to have multiple "open" positions that work with different keys, either different ones or a standard and a master key. Regards, Walter.
[cctalk] Re: Philips P3800 key lock swtch
> On Apr 8, 2025, at 12:06 PM, Tony Duell via cctalk > wrote: > > ... > I think there is only one key. The lock is somewhat unsual though. > > There are 4 pins in the plug (the rotating part). There are 3 > positions of the lock and the key is removeable in all 3. There are > therefore 3 sets of upper pins and springs in the housing. The 3 > positions are 'OFF' (anticlockwise), ON (centre), PM (clockwise). The > visible holes in the housing were always empty, I've not removed > anything there. I think they were there to allow the springs and upper > pins to be inserted when the lock to be assembled. Note that the hole > for the 'OFF' position are ofset away from the front of the lock. > > Here's what I think happens. In the ON position you can use the key > and simply turn to PM. And go back again But to select OFF you have > to push the key and plug inwards before turning the key. To turn on, > you just turn the key (you can't go too far and select PM by > accident), the plug will then spring forwards. There was a compression > spring in the back of the lock, forcing the plug forwards. I have > removed this now for the moment. > > The in/out motion of the plug is controlled by a pin (the end of which > is the silver circle in the 'side' photo) running in a groove in the > plug. That all makes sense. To allow removal of the key there have to be holes at that position to let the pins (the ones in the cylinder) move upward as the key is extracted. Then normally the pins in the lock body would be pushed inward by the springs, locking the cylinder. If for some of the positions you have holes but they are empty, it means the key can be removed in that position, and then the pins would probably drop back into the cylinder, leaving it unlocked and free to turn. But in addition, you could turn the lock upside down and let the pins slide out (be sure to keep them in order). The lengths of those pins would tell you the cuts needed for the key. paul
[cctalk] Re: Philips P3800 key lock swtch
On Monday, April 7th, 2025 at 12:24, Paul Koning via cctalk wrote: > > https://yagi.h-net.org/philips_p3800_lock/ > > It looks like a Lips lock, which would make sense for a Dutch computer. I had some parts of me go looking around, and the keyway reminds me a bit of an EVVA Euro cylinder, per here: https://www.art-of-lockpicking.com/paracentric-keyways/ So, looking at those it seems like five-pin Euro lock blanks would be a good thing to look into. The Doctor [412/724/301/703/415/510] WWW: https://drwho.virtadpt.net/ Get thee down. Be thou funky.
[cctalk] Re: Philips P3800 key lock swtch
On Monday, April 7th, 2025 at 13:05, Fred Cisin via cctalk wrote: > although time consuming, and requiring som moderate machining skill, that > looks like it shouldn't be an extreme problem to make a blank. > I assume that Tony has, or has access to, a precision milling machine. It might not even need an automill. At the very least for the purposes of prototyping, it should be possible to carve some of the wards by hand with a Dremel. I've done that for a couple of mechanisms I've restored over the years, and if you don't mind taking two or three tries to get one that fits right it's doable. > The rear picture shows that it is not a fully paracentric keyway. (where > "ridges" and "valleys" of the milling of the keyway would cross the center > line, making it impossible to insert a very thin flat piece of metal) Ooh - good catch! That's a really good point. The Doctor [412/724/301/703/415/510] WWW: https://drwho.virtadpt.net/ Get thee down. Be thou funky.
[cctalk] Re: Philips P3800 key lock swtch
On 4/8/25 21:14, The Doctor via cctalk wrote: Would it be feasible to cut a couple of key blanks out of metal plate? Maybe metal strips? (I'm a CAD designer, have experience in machine shops and have done custom metal fabrication off & on for almost 50 years) Entirely feasible, and expensive. I had only been following this thread casually, but I think you have the necessary information in hand. (I'm gonna go back and reread the thread; now I'm reeled in) You need precise (to your spec of precise - I don't know how much "slop" is too much for a key) measurements of the overall thickness and width, length, and the depth and width of each groove. I'm inferring that you're not certain whether one key performs all functions or several keys with different groove patterns will be needed. If I were setting up for the latter, I would want all possible configurations in hand before starting. Fixturing and securing the raw material is a big part of the expense, so I would want the fixtures - jigs - to allow for all needed patterns. I have the software and the machinery needed to do this, and if I were on your side of the water I'd love to be involved, at no cost. It's a really interesting task. Doc
[cctalk] Re: Philips P3800 key lock swtch
On Monday, April 7th, 2025 at 02:07, Tony Duell via cctalk wrote: > However the key/blank is a rather odd shape and I have nothing that > will fit in the keyway of the lock. How likely do you think it would be for someone to machine the right kind of blanks out of stock? Or possibly existing key blanks that just need a few extra cuts on the sides? The Doctor [412/724/301/703/415/510] WWW: https://drwho.virtadpt.net/ Get thee down. Be thou funky.
[cctalk] Re: Why I am not worried about AI taking over the world!
Nay-saying is always easy. I'd rather hear from smart people who can tell me whether or how a tool is useful now, as well as how it might enhance projects in the future. Why couldn't a present or future AI help you translate from one BASIC dialect to another, if that's what you want to do? If you want a tool to do that, how could you help make it possible? How can AI tools help with disassembly? How might they help you search and understand and get answers from all the files in your digital packratting? For the record, my first paying programming job in '81 was translating between dialects of BASIC. - John It possibly could! Try it with ChatGPT and see if it works. For an example of how bad ass the AI stuff is A friend and I are supposed to supply a bunch of restored pinball and arcades to a bar restaurant arcade type thing. Vendors are asking around $10,000 for a change machine that can take credit card payment and dispense tokens. The people putting the place together asked if machines can be had used. I tell them, for dollar bills easy but not the modern credit card ones. Can I make one, they ask. I start digging around looking for info on those terminals and how transaction setups work for them to judge if it would be easy to add a subsystem to an older change machine to accept card payments and dispense tokens. Developer only portals that require signups, fake sites trying to harvest data and sell you their services, sigh. Google used to be better. Hit up ChatGPT. A few specifically asked questions and it craps out example code for telling the exact terminal I asked about the amount you wish to charge and picking up status. Very cool, and more than I was looking for. Confirms that the card info staying on the PCI compliant CC terminal and never crossing to my system should meet PCI compliance. Asked it more questions about payment gateways, processors, and is it possible to share those accounts between popular touchscreen point of sale terminals. The thing answered a lot and gave a lot of specific examples of various brands of equipment and if you can share the accounts. All in like 5 minutes. I could get it to crank out the code for the end user UI as well as code for arduino to manage communications to the 90s era change machine easily. It just reduced a bunch of web searches to 5 minutes with very specific examples. Some risk it could be wrong, but nothing major. It's wild, give it a try. ChatGPT is free. Throw it some odd stuff, and give it quite a bit of detail in your question. I just asked it to write me code for IMSAI 8080 to bounce LED back and forth and it is cranking it out with comments in the code. It tells me about the front panel switches, but if I have a monitor rom it's easier. Asked me if I want instructions on toggling it in or data for a rom monitor even. At VCF East someone griped that these GPT things are providing data that is stolen from other people. Be it the artwork or the code or whatever. I kind of fail to see the difference in training it on a large data set versus some kid learning to play piano by playing the Thexder theme or Arbands books for brass or whatever. We all build on what have learned from other people's examples. It's neat. -- : Ethan O'Toole
[cctalk] Re: Philips P3800 key lock swtch
On 4/8/25 21:41, The Doctor via cctalk wrote: I don't know where the unit is, and I'm in the States. If you'd be on the west coast at some point I'd love to geek out over metalworking and lock parts over coffee some time. I'm deep in the Texas boondocks, but Skype is a thing. :-) I do hope y'all keep posting progress here. I know it's real peripheral to the list's interests, but it's not like we're *that* busy... Doc
[cctalk] Re: Philips P3800 key lock swtch
> How likely do you think it would be for someone to machine the right > kind of blanks out of stock? Or possibly existing key blanks that > just need a few extra cuts on the sides? It's really not going to be necessary to machine blanks. Download "Key Blank Directory - Edition: 13" from here: https://www.ilco.us/support/resources and explore e.g. Yale keyways. The lock may not have been built by Yale, but that approximate keyway shape is widely used. De
[cctalk] Re: Philips P3800 key lock swtch
On Tuesday, April 8th, 2025 at 00:17, Adrian Godwin via cctalk wrote: > They do run a LockCon but it appears to be in October and costs around 360 > Euro plus accommodation. So, an expensive route to find a key. > The UK organisation run a monthly meeting at FizzPop in Birmingham which > is more accessible but perhaps less likely to recognise the keyway. Just reach out to them - or TOOOL.us if you're in the states. Both arms of the organization have chapters all over the world. Most of us are pretty old-school in that we still like e-mail lists so a few folks might join this list to help. :) The Doctor [412/724/301/703/415/510] WWW: https://drwho.virtadpt.net/ Get thee down. Be thou funky.
[cctalk] Re: Philips P3800 key lock swtch
although time consuming, and requiring som moderate machining skill, that looks like it shouldn't be an extreme problem to make a blank. I assume that Tony has, or has access to, a precision milling machine. On Wed, 9 Apr 2025, The Doctor via cctalk wrote: It might not even need an automill. At the very least for the purposes of prototyping, it should be possible to carve some of the wards by hand with a Dremel. I've done that for a couple of mechanisms I've restored over the years, and if you don't mind taking two or three tries to get one that fits right it's doable. I am guessing that Tony has enough skill milling, that a dremel would not be much quicker or easier. Besides, depending on what kind(s) of milling machine that he has or has access to, it would be straight forward for him to mill a LONG piece of brass stock to fit. Then he could cut short pieces for the key blanks. The rear picture shows that it is not a fully paracentric keyway. (where "ridges" and "valleys" of the milling of the keyway would cross the center line, making it impossible to insert a very thin flat piece of metal) Ooh - good catch! That's a really good point. If actually milling or "dremel milling" the grooves, paracentric and non-paracentric would not differ much. BUT, if it is non-paracentric, then a thin piece of flat brass stock coule fit into the keyway without necessarily haveing to mill alll of the grooves and/or not having to mill them deeply. -- Grumpy Ol' Fred ci...@xenosoft.com
[cctalk] Re: Philips P3800 key lock swtch
On 4/8/25 21:14, The Doctor via cctalk wrote: Would it be feasible to cut a couple of key blanks out of metal plate? Maybe metal strips? Meant to add that if you go this route, a dozen of each blank shouldn't cost much more than one. The cost of making them is mostly in the design and the machine setup, so be sure to get as many as you might need on the first order. Doc
[cctalk] Re: Philips P3800 key lock swtch
On Wed, 9 Apr 2025, The Doctor via cctalk wrote: My thinking was the former - one key cut for five pin stacks, but only four would be used by a given position. Tony said that the plug slides forward and back, with a spring behind it. With that setup, four cutes is likely to work, since you slide the key and plug further back for the offset position. As you mentioned, with a five cut key, but with different sets of four pins, then it should be able to work WITHOUT pushing the key and plug further in. In terms of how much slop there can be, . . . most lock companies (at least Schlage and BEST) make some keys that will fit (a little more loosely) multiple different keyways. Those are called "multi-milled" or "multi-section" Schlage SC20 fits C, CE, E, EF, F, FG, and G. Whereas none of those keys will fit any of the others, or vice versa. similarly, fitting fewer keyways, Schlage H will fit C, CE, and E Schlage J will fit E, EF, and F Schlage K will fit F, FG, and G But, the "L" or SC20 fits all of those; it's a thinner key, and is a little sloppier than using the "right" key. and, yes, those are far from all of Schlage's keyways. BEST FM? and JKLM? -- Grumpy Ol' Fred ci...@xenosoft.com
[cctalk] Re: Philips P3800 key lock swtch
On Tuesday, April 8th, 2025 at 19:30, Doc Shipley via cctalk wrote: > You need precise (to your spec of precise - I don't know how much "slop" > is too much for a key) measurements of the overall thickness and width, > length, and the depth and width of each groove. I'm not a machinist (definitely not with your experience) so I don't know. To be honest, just as "open is open," "it fits is it fits," and if that involved a little work with some rattail files to make sure the blank fit, that's fine by me. Your mileage may vary, etc. > I'm inferring that you're not certain whether one key performs all > functions or several keys with different groove patterns will be needed. My thinking was the former - one key cut for five pin stacks, but only four would be used by a given position. > I have the software and the machinery needed to do this, and if I were > on your side of the water I'd love to be involved, at no cost. It's a > really interesting task. I don't know where the unit is, and I'm in the States. If you'd be on the west coast at some point I'd love to geek out over metalworking and lock parts over coffee some time. The Doctor [412/724/301/703/415/510] WWW: https://drwho.virtadpt.net/ Get thee down. Be thou funky.
[cctalk] Re: Philips P3800 key lock swtch
On Wed, Apr 9, 2025 at 4:07 AM The Doctor via cctalk wrote: > I don't know where the unit is, and I'm in the States. If you'd be on the > west coast at some point I'd love to geek out over metalworking and lock > parts over coffee some time. The lock (and I) are in SE London, England. I think travelling over 6000 miles is a bit far :-) -tony
[cctalk] Re: Philips P3800 key lock swtch
On Wed, Apr 9, 2025 at 3:22 AM The Doctor via cctalk wrote: > Before I saw the photographs I would have guessed that the three sets of > tumbler > stacks would all be pinned identically. In terms of manufacture and > maintenance it > would be both easiest and cost effective. But. > > Photograph number five on that gallery page (nice and simple - I like it) > makes me > wonder. To have a key that would work with all three sets of tumblers it > seems > like it would have to have five cuts on it, not four. The four cuts closest > to No, I thnk it's just 4 cuts. I am pretty sure I can make out 4, and only 4 pins. In each of the 3 positions those line up with the appropriate 4 upper pins. The plug moves back for the anticlockwise position. I must admit I had the same thought initially of 2 key types but I can't see how that would work. > the shoulder would work in the (looking at the lock straight on from the > front) > rightmost and top positions. There should be room in the keyway for the > unused fifth > cut at the end of the key, with no pin stack to engage. However, the leftmost > pinstack is offset backward by a distance of one pin stack, which implies > that it is > pinned for the second, third, fourth, and postulated fifth cuts on the > matching key. > > It would make sense to find key blanks that almost but not quite fit - with a > blade > of the right width and height but ideally without any of the warding grooves > on the > sides (but that may not be feasible). Then grind out wards that match the > weird > parts of the keyway. > > But that might not be practical, either. Using a 3d printer to fab a couple > of key > blanks wouldn't be a good idea, I think. I don't think plastic would hold up > well. It wouldn't (and in my limited experience 3D printed plastic parts are weaker than plastic parts machines from solid anyway). It takes considerable torque to turn the camshaft in certain positions. > > Would it be feasible to cut a couple of key blanks out of metal plate? Maybe > metal > strips? Should be possible to mill them, but as I said I'd want to make several identical ones which is a pain without CNC tools (my lathe/mill are entirely manual). I can't believe it's a custom keyway so blanks should exist for it. -tony
