Toastie

Yeah, fabricated in these simply breathtakingly expensive molding machines, that potentially can take TLG down if they would go nuts! I wonder who is behind these companies ... or ... wait ... are these molding machines - not that expensive at all??? Who knows ... Best Thorsten

I second that. The review is absolutely >perfect<, no criticism - other than the fact that this set was sponsored. Believe me, I know what it means to make enough money to run any kind of joint. My joint is a research group, depending 95% on extramural funding. And industry plays a major role in doing so, as available federal funds are declining significantly - waging/participating/suffering from trade wars and/or real wars is always the stupidest thing to "happen". But: As we have a dedicated competitor thread on EB, as well as a Community Forum, it would be fair to mark any reviewed set (usually very high rankings, often not so much shared in the webs) with "Set sponsored by TLG) - if it is sponsored that is, of course. If someone just purchased it, reviewed it, and put it up here: Fine with me! There is of course >never ever< any other than a LEGO/TLG set review on the front page (duh), but it would be fair to know: Sponsored or purchased ... As we all know, that issue drives some folks out there crazy when they watch reviews on-line. Well, I am a rather relaxed with all that. For me, TLG's present stuff has moved to my "WHAT???" list. The items of the old stuff is on the "WANTED" list ... All the best Thorsten

Repeated quote What about Cobi/CaDA/etc. are simply not caring about TLG copying? They just make new designs (or whatever it is called in the legal world), and sell sets using it. There are two things (as far as I am concerned): When a) TLG uses the stolen designs in their own sets, they are as bad as any other "knock-off" and that could be nicely thrown at them (but no die hard would care, I know) and b) maybe having far, far less heavily overpaid legal superstars on their payroll saves them a sh*t load of money so that they can hire more creative and well-educated designers, who simply come-up with new designs at a pace that keeps TLG busy with copying. I am kidding, of course. On second thought though ... no, I am kidding. Best Thorsten

As well as these ... https://www.eurobricks.com/forum/forums/topic/7167-brick-de-yellowing-techniques/ https://www.eurobricks.com/forum/forums/topic/57827-yellowing-in-misb-sets/ https://www.eurobricks.com/forum/forums/topic/97365-lego-and-sunlight/ https://www.eurobricks.com/forum/forums/topic/171633-the-old-yellowed-lego-returns-to-shine/ https://www.eurobricks.com/forum/forums/topic/199644-mindstorms-nxt-retrobright-easy-technique-fix-your-yellowed-nxt/ ... and even more. There is one thing, though, I'd rather remove from any "why" discussion: Bromine. That is certainly not the case, as has been discussed elsewhere. It would never work; if bromine as brown molecule would be present in the outer layers of ABS accessible to light/chemistry, it would react away within no time. Everything else (light, heat, other chemicals, surface damage, polymer formulation etc.) is up to speculation. It for sure is a "close to the surface effect" as you can sand off the yellow @MAB has said. It does work, but removes material ... Best Thorsten

I was waiting for such a reply, before beginning bragging ... that result indeed is very interesting (!) and contradictory to what I wanted to ague. On the other hand, I really believe all that yellowing depends on so many (not controllable by us) parameters, that any global prediction is rendered useless. However ;) statistically (with some sort of narrowed error of margin), I'd argue: Colored ABS yellows (white) over decades (years), because some unsaturated bonds in the ABS polymer break, shifting the absorption spectrum deeper into the visible region. White brick = close to 100% reflection, yellowed brick = loss of refection within the visible (VIS) spectrum. It is not necessarily the presence of free electrons in the polymer, but much more so "saturation" of the initial radical sites by reaction with another ubiquitous biradical always present, the oxygen molecule O2, forming oxidized sites in the polymer. Which in turn shifts the absorption spectrum into the VIS region. Upon H2O2 treatment, these sites are oxidized again, and a "temporary" effect is observed. However, I believe H2O2 is not selectively attacking the culprit sites, it also bites into the saturated ABS polymer sites, generating free radicals where there should be no free radicals generated upon exposure to light and such. Which in turn accelerates the yellowing processes, as now, formerly passive saturated bonds are activated. All total and utter speculation, but I am still favoring the radical mechanism (found in the literature, along with many others) as the main cause/driving force for yellowing. This proposed "mechanism" depends heavily on the parameters a) degree of polymerization (= yield/creation of single bonds) in the original ABS polymer, b) exposure time to actinic light, c) change of surface layer composition (grease, water etc.), d) change of surface constitution (scratches, dents, etc.), and so many more. And H2O2 may lead to temporary shifts, but in the end, induces sites accelerating the yellowing process. As said, total speculation. Do I know any other means of “whitening”? Yeah: White (spray) paint. But that is generally not accepted at all. They for sure do it in the LEGO Lands, as I and others have literally observed it. All the best Thorsten

Oh, how lovely is that! In that case, I would make 15 PCB's for the committed cards (which would further shrink the size of the "add-on" significantly. Then I have to order a passive ISA-bus extension back plane, the ISA bus connectors, and then check it out. As I have total confidence in Jo's suggested 74LS85/4bit switch approach used on my perfboard, how about a lill' punt? If it works, I get 2 of @evank's 9771 cards, and >extended< bragging rights on his website? And if it fails, then restore all cards to original configuration, and sacrifice my IBM XT to the Gods of Silicon? Just to clarify: It is all out there, this is no hollow promise: If this works (reference thread and picture source: https://www.vogons.org/viewtopic.php?f=46&t=41599) with 13 ISA graphics cards, it will work with 16 9771 cards for sure: Source: http://i.imgur.com/40ecbL4.jpg So world, send me your 9771 cards for an upgrade, you have been waiting for almost 4 decades! All the best Thorsten

Dear all, I like to announce that anyone in possession of 16 LEGO 9771 ISA cards can now build a 4.5V project operating 16 x 6 = 96 motors and 16 x 2 = 32 sensors and use them from within one program. All you need is access to the ISA bus via port addressing. I know at least one person who thinks along these lines, as bigger is better! So, here we go, @evank, @BrickTronic, @alexGS, and all other TC Control freaks First, I like to show some photographs of the original 9771 that I got from a very friendly person here on EB, essentially for free. The photos I found in the webs were a bit blurry, but I may not have searched long enough. All pictures are also here: https://bricksafe.com/pages/Toastie/9771-isa-card Here is the "plan", based on @BrickTronic's proposal to use a 74LS85 (4-bit magnitude comparator = 16 addresses resolved) instead of my idea (74LS238, 3 line to 8 line demultiplexer), which would have required an additional wired OR gate + pull up/down resistors = 8 addresses resolved), as illustrated in the circuit diagrams shown a few posts further up. I also followed Jo's advice and directly connected IC6/3 to +5V. 4 trace cuts as indicated are necessary - they are easily “repairable” should the card be restored to original configuration: Here is the real thing with the additionally soldered wires connecting to the little perfboard via a Dupont plug and socket. The original configuration, listening to address 925 (0x39D) with closed “address switch” bridge or 926 (0x39E) with open bridge, is restored by replacing the perfboard with the plug shown on top of the interface card. Note that no further harm (other than - reversibly - cutting four traces on the PCB. The perfboard is fastened using the original hole in the 9771 PCB with a plastic screw, brass nut, and rubber washer, so it hardly "wiggles". No glue applied other than for fastening the additional wires, again, fully reversible - simply apply "hot glue heat" for removal: This is the perfboard; the switches are resembling 1101 = 13 (on = GND = 0; off = internally pulled up in 74LS85 to +5V = 1) at base address 1110010000 = 912; 912 + 13 = 925 which is the original address as used in TCLogo: The address range is thus from 912 (0x390, all switches ON = 0000) to 927 (0x39F, all switches OFF = 1111). This range happens to be the same as used in the IBM XT (5150/60) for the IBM cluster interface ISA card. This card is a predecessor of the token ring or Ethernet cards, which came later. The cluster interface allowed to hooking-up 63 XTs (IBM "cluster") to one XT file server. Should you plan on that, then sorry, the approach shown here won't work for you . Some notes: TCLogo only works when using address 925 (0x39D). With QBasic or the like, all addresses work, when using "OUT address, 6 bit pattern", or "IN address, variable". The 74LS85 manages the 1 kHz PWM pattern (always) applied in TCLogo, see posts a couple of pages ago, easily - well it should, as the LS chip can do the comparison in about 25 ns = 40 MHz (https://www.ti.com/lit/ds/symlink/sn54ls85.pdf?ts=1761490286532). The power levels in TCLogo go from 1 (minimum) to 7 (maximum). In this photograph, the leftmost lamp is running at power level 6, the rightmost at power level 1. 4.5V Lamps are not necessarily the best power meters, but one gets the idea: Here is a photograph showing the setup: And this is my first TCLogo program. It took me hours to figure out how to increment the value of a variable using the LEGO instructions for TCLogo. Google's AI told me how to do it: MAKE "variable :variable + 1. Which now sounds perfectly OK. I tried so many other things ... And I know that there are ten better ways of doing this in Logo, but hey, I could have printing "Hello world" on the screen as my first Logo program ... Note to Jo: The IC labeling, e.g., IC1a, although there is no further gate in 74LS30, is caused by the program (Target 3001, free version PCBPool). It always assigns an "a" to any IC, regardless of its interior. Sorry for that, but I am not an electrical engineer and just happy that I sort of know, how to use Target 3001. I bet there is a way to tell Target not to do it this way, but I couldn't figure out how ...) All the best, Thorsten

Jeez - that is >so not< an "original" R5 - wife and I drove one (Model 1974, with crutch shift) for years. It had less than 10000 km on the tachometer when we got it from grampa in 1983 (for free), and I kept it up and running for almost a decade. to the real R5 Best Thorsten

100% agreed. Another thing is: We are talking about 20 to more than 30 years (!) of lifetime of a chemical formulation ("glue" or adhesive) that holds together two rather smooth surfaces of different composition: The sticker material and ABS (the bricks). In many instances, the ABS surface is not ABS, but a mixture of everything that touched it before the sticker is applied, including fingerprints ;) The glue on the sticker sees the “free environment” for the first time when it is peeled off the sticker sheet, and is thus rather clean when it is applied. Provided of course the glue surface was not touched, the sticker was applied in one go and not realigned or even peeled off and applied again, because the first try was intolerably bad, and so on. Which may heavily impact on the lifetime of a sticker. And then the countdown begins: Dark = good, humidity depends on sticker material/glue formulation, temperature not too hot, not too cold, and so on, and so on. The sticker material certainly changed over the decades, as did the glue. I have stickers sitting on #727 looking rather good, maybe the color fainted a bit, but what do I know about colors. Definitely more paper-like sticker material. In other words: They may suffer from high humidity and sweaty fingers much more than today's thin plastic material, which in turn falls apart after a couple of years to decades upon environmental exposure. Generally, I guess stickers are not designed to live for decades, but rather “look good” and “withstand” repeated finger massaging (while glued to the brick) for a couple of years. Or months, possibly hours, depending on the attention-span and interest. I also believe that a lifetime pattern (which sets/decade had the more durable stickers) is really tough to identify, as there are simply too many variables in play ... But who knows. TLG would ... Best regards Thorsten

Moin Sven, oh my, oh my ... I am not a car person, but after seeing this, I may reconsider that. What a beautiful and >perfectly< designed and assembled replica of that Benz! Crazy. I looked for a while at the last picture, and then it made click. Of course, you couldn't leave out a perfectly designed “underside” as well. So when you show off this beauty in a diorama, it can very well go into a glass box, which slowly rotates around its main axis ... 360° of perfection. Wow. All the best Thorsten

So do I. I am missing some spiritual liquids All the best, and let the maximum entropy in the system rule! Thorsten

OK. Honestly? Your list is way too short: The entire appearance of the locomotive? The simply >perfect< match with the original? The vivid colors? The choice of the original? The angles? The colors? This is so incredibly well "done" - "created" is much better - I am in total awe. Congratulations, and thank you very much for sharing. Best wishes Thorsten P.S.: If these are not full quality pictures, I am thanking Flickr - otherwise I may have fainted.

Hmmm - I believe they do know how to do it properly, it depends on the perspective, doesn't it? As far as I am hearing people discussing this issue, Technic as it is/as the sets are designed right now sells/sell very well. For a for-profit "organization", that seems to be the proper goal. I believe every other company out there does it the same way: If you can sell way more of would-be-Technic-sets with a Technic tag slapped on them, then why make true-Technic-sets, even a few, with a Technic tag slapped ion them? (Of course I know why one should do that: Because this would be the true spirit and idea of Technic. But spreadsheet programs having a column "profit" rule.) TLG is just another of the large businesses out there. Best Thorsten

And I find that absolutely wonderful! And really appreciate this. This is not that common anymore. You put a lot of thoughts into your project - from the electronic design stage all the way through the customizable case! There are a couple of "things" (let's call them simply "things", not wishes, issues, or dreams; these are naturally diverse, every individual here and elsewhere has individual demands). For one, you are targeting PF and, with a conversion cable, 9V folks. PF and 9V is phased out since long. So folks get their supplies from the BL aftermarket - or from China. The latter is definitely cheaper. Once you take that route, some call The Dark Side (which to me is utter nonsense: The moment I can't get an original part anymore for every single device here in the house, I go China, believe me), you all of a sudden feel "free". At least this happened to me - today I can drill into any LEGO piece if I want to, because TLG did not make what I need, without getting nightmares or any bad feeling. It is ABS I am drilling into. At that point, the advantages of any newly developed third party device designed for LEGO stuff but targeting the LEGO aftermarket must be compelling. Otherwise, it will be just: Naa, can get this and that here and there for less. See, what I would do, if I required a nicely controllable 9V DC power supply? I'd go with a modern Chinese lab power supply, 0 - 12 V, 0 - 5 or 10 or more amps. The current limitation is good for testing purposes or simply "early motor burn out protection". 9V (train) regulators don't need proprietary power supplies, this is one of the many TLG myths. Or lies. Anyone in the correct range will do. The 9V controller is internally nicely protected, nothing much can happen there. No, the usual stuff won't work, as Pete Venkman said in Ghostbusters, when he tried to get in touch with hat lady-ghost in the library. Not the PF terminal, not the amperage, not the power supply. Your approach has this powerful extra, though, a microcontroller. ESPs are crazy. And dead cheap if you skip the Arduino part. With WiFi, BL, BLE, easily connected to an (OLED) display, it has many inputs. Which makes a total difference. The thing is: a) How to use it for b) what purposes. Do GBC folks need that? I don't know. Do train folks require that? Hmm, maybe? Do Technic (machines) need that? I think so. But then you need inputs. You said you want to do loops per programming. Well, there are not that may static/constant loops, there are controlled loops. I have no l clue how to get sensors into your project, but it would make such a difference: A smart >power supply<. Not a smart brick, they are out there: A power supply. I, for one, like the display. Others will tell you: I can still "read" a dial (which is true, I can as well, but). I like the wireless stuff. Others will tell you: Go to a show and BOOM, you can hear Radio Gaga, but not reliably control your contraption. So what, I don't do shows, I have fun@home. Hmm, as said, I love your project. I find it absolutely original, fun, and appealing. See, I make my own stuff, virtually >nobody< here is interested in, except a handful of total nerds (= wonderful people), which share the notion of "just sharing a project I made and putting it out there for the community". These are your words. And THAT is totally cool and what drives me. I will closely follow what you come up with! Keep on rolling!!! All the best Thorsten

It does, here is how to activate it - it is a bit "hidden": https://medium.com/@slovati/esp32-a-look-at-the-built-in-basic-interpreter-4b15725457b And it is fully in line with @evank's wish to use line numbers: You have to when programming. No line numbers and commands are executed directly - as on any 8 bit wonder ... Happy POKE'ing! Best Thorsten

I'd say so. It happens, it is normal, kids, or grown-ups running the motor ;) They do wear out. The thing for me always is: What can be done to get the rest (i.e., everything else than the motor) back to life? Which is a replacement motor. There are so many out there, and it'll need a little research. I, for one, would try to do it right away. Just for the purpose of >not< leaving the unit alone and, in the worst case, trashing it. Soldering is not rocket science, particularly if you need to remove a wire. My view: It is a learning curve. Desolder, inspect, Google. What is the voltage on these devices (I am a nerd, do only lame LEGO stuff, so don't know anything about racers or the like)?. Then the footprint? Seems to be a generic type of motor. Dimensions? And maybe others do know more about that motor here on EB? Actually, I bet they do. Maybe this forum is not the most appropriate, Technic maybe? In any case, just keep providing data and we shall try. All the best Thorsten

Wait - including the brass gear sitting directly on the axle of the motor? Or is the motor axle not turning easily when detached from the other gears? Best Thorsten

I am at it - thank you very much for your support! I really appreciate that. Will do as you suggested, Jo! All the best and have a nice weekend Thorsten

Oh yes - I am writing such words (and in public) only to a very good and deeply trusted friend, believe me! Alright, challenge accepted ;) OK, I have the 5160 XT (the IBM AT was a 5170, wasn't it?) I need the floppy controller and graphics card to a bare minimum. That leaves me with 5 cards, as I can't use slot 8 with any of these cards, including 9771 (https://www.minuszerodegrees.net/5160/misc/5160_slot_8.htm). Let's now assume, I'd have the expansion card for 5161 (which cannot go into slot 8 of the 5160 XT either), a 5161 expansion unit, and the expansion card that goes into 5161. That leaves me with 4 slots in the XT. In the 5161 expansion unit, the extender card goes into slot 8 (https://minuszerodegrees.net/5161/misc/5161_basics.htm), which gives me 7 free extra slots. 4+7=11. In other words, my minimally invasive address decoding addition to 9771 cards needs to resolve 16 addresses to beat @evank's record - and all that is totally and utterly fictional!!! However, I love such nonsense. Totally. It is fun. And this is why I will use @BrickTronic's proposal to do so Evan, please keep us updated on Blocky Kong! This thing is a blast. I truly love it. All the best Thorsten

Shoot. Can you turn the brass gear manually when lifting the motor away from the remaining gears? If it does not, then it seems to be the motor, if it does, then further investigation (e.g., apply an external power supply to the motor after de-soldering one of the original wires) seems to be warranted. Best Thorsten

Whoa!!! That's an announcement! So, Interface A had 6 outputs and 2 inputs. Interface B had 8 outputs and 8 inputs. Interface C has 1 output and no input? Seems to me more like a smart power supply rather than an interface? "Smart" as in providing Wi-Fi/BT/BLE/MTQQ/etc. connectivity and control (well, all that an ESP32 provides). This should also allow for control via home-integration/assistants. So yes, it is an interface! And very nicely done - it has a really appealing look. Congratulations on your achievements!!! Bringing such a device to the market is really something. With that USB port connected to the ESP32 hybrid controller inside (not only for diagnosis, but power-programming!), along with an X-channel USB switch, you can also orchestrate a nice X-channel Interface. Sensors can be hooked-up wirelessly. It will be a little pricey though - well interfaces A and B were sure as well! All the best, and thanks for sharing! Thorsten

Hmmm. There is this little screw; when you remove that, can you get the motor out? It seems it is connected with cables to the PCB. If you can get it out: Does it spin without any load? If not, then it may be the motor. Next step would be to learn more about this motor ... Best Thorsten

As I said, I'll be back ... @BrickTronic Jo, "quick" question - do you think, this could work? I just ordered a couple of 74LS85's - these are really powerful! Specifically: The B inputs are floating, when a corresponding switch S is not closed. Well, as far as I remember, LS chips are quite stable regarding open inputs as they are internally pulled high. If not, then I would need to add 4 pull up resistors to the B inputs, right? Also, an address is inversely selected, right? All S closed = 0000 (0), all open = 1111 (15). Correct? Thank you very much for pointing me to 7485, a really enjoyable chip! (I know, the nerd I am. It's OK ) All the best Thorsten NOTE: For the first time, IC6b, pin 6 connects correctly to IC1a, pin 11 (and not 12 as always before in this thread)

EDIT: Go to the second quote below: Jo has just >>sunk<< my approach described above. Yeah, why not using 74LS85? I tell you why: I had never heard of it, just looked at its function table a couple of minutes ago, had to read Jo's question 5 times, and finally, got it. This is >so< elegant!!! First: Thank you very much Jo for checking!!! I trust your assessment very much, you know that. Second: 74LS183 would be "easier to use" as I have them floating around here. I had exactly scribbled down what you described. Nice. And yes, using the now obsolete IC3a/b would make nice inverters. However, here is more about my choice of 74HCT238 over 74LS138: I'd like to be able to "restore" the 9771 card to its original design/behavior - for no obvious reason than - I want to. Don't ask me why; I am trying to do that with all the "ancient" machines I am tinkering with. It is simply my respect for the original designers and hardware layout - I know it sounds absurd, but so is the entire world of today All I need to do is cutting 3 traces on the 9771 card (>tiny< cuts with a razor blade ) and then solder >tiny< loop back wires to a female DuPont connector I am planning to reversibly hot glue to the enormous empty space area of the ISA card. If I now plug in a male DuPont connector with 3 bridges, it should work as before on addresses dec 925 or dec 926. For the new address range 920 - 927, I'd remove the male DuPont connector and replace it with a little perf board, carrying the 74HCT238, 8 bit DIP switch, 8 diodes, and two 1k resistors. To make it work, I have to additionally wire A0, A1, +5V, GND in parallel to the female DuPont connector, which are unused in original mode, but populated when using the lill' board ... And why 8 addresses? Just to theoretically outpace @evank's Apple ][ setup; he uses up-to (I believe) five 9767 cards ... The difference would be: My hardware addition only works on an IBM XT system with the 5161 extension box with 8 cards. But I don't have 5161, nor 8 9771 cards. It is just for fun, that's really all. I am busted, though, when he finds an Apple expansion board ... BUT, be aware folks, I am prepared to go to 16 cards should that happen - I shall then just swap 74LS238 with 74LS154! As I said: More crazy stuff to come. Because you are a brilliant genius, Jo, and I am a lousy chemist pretending to know a bit about TTL circuits. This is insanely elegant, and I will certainly do exactly that! Let me draw up your proposal! My original approach using the DuPont loop back connector and the additional lines A0, A1, +5V and GND lines still works as intended, but I just need one 74LS85 and a 3 bit DIP switch. Need to order the former, have the latter. And the best part is: "These devices are fully expandable to any number of bits without external gates", says the data sheet. Thanks a million, Jo, you made my day!!! I will be back ... Best regards Thorsten

Wait — does that mean, you can charge the Cube while it is, e.g., running a motor? I never checked that on my Cubes. This would mean, that you can charge them on the fly, is that so? Missed that as well, nice! Well, I am using the Cubes always in combination with the LEGO PUp remote (and an ESP32 “in between”). Thank you for the heads-up! Best Thorsten