Toastie

Wow, that is a nice action scene!!! I really like this scale, as I find it to be most challenging. You captured that scene perfectly (I like the angled buoy very much); the red/white border is cool as well. And the boat, of course - pumping out that water fountain! Thank you for sharing! Best Thorsten

It is a PTC (thermistor), isn't it? Seems to be in series with the lead as it is for the M-motor (https://www.mikrocontroller.net/topic/267085). Yes, I know it is in German, but they conclude that the M-motor has a PTC in series. As all the other TLG motors have. Your measurement shows R = 1 Ohm. At 500 mA, this thing needs to dissipate P = IR^2 = 0.5W, which will certainly heating it up "a bit" - which will increase the resistance, reduce the current. As R scales quadratically with P though, there is always a chance that such a device causes trouble, when running off from specs. As R(T) of the PTC goes through the roof at lets say 80°C or so (and then essentially shuts down the motor) there is this "in-between" region ... This is interesting! Best Thorsten

Before you go there: @dr_spock's post is the way to go before using invasive methods. First a good diagnosis, then on to the blades, knifes, saws, Bunsen burners, etc. Do you have a meter? Resistance between C1 and C2 is of interest - also when working that cable. I am sure you have seen this: http://www.technicbricks.com/2012/08/tbs-technuggets-13-inside-pf-l-motor.html With reference to this article, on the pictures of the opened motor, there is this one piece of additional electronics - it could be a thermistor, but I can't tell, as I can't zoom in nor is it clear where the wires are going. Should you take the surgery approach: Simply electrically shorten it, then let the motor run under some load (if it turns at all) by giving that axles a little work to do (two fingers ... or whatever). As a rule of thumb: First dies the thermistor, and the motor is fine. Best Thorsten

Oh yes. Let's see: I was 15 when I was there for the first time. Then basically every year. At least for the past 20 years. Even when we lived in the US, more than 20 years ago and visited Germany for X-Mas we spent 5 days there over New Years ... I'd live there tomorrow ... but a) Germans can't buy (good thing) and b) they simply don't need chemists there, well at least none of the kind I am. So we have to compensate for that with vacation time ;) All the best Thorsten

Ha, flooding this thread with - most probably - irrelevant info! Well: Whatever So, here is a link to the ZX Spectrum BASIC code: https://brickshelf.com/gallery/ThorstenB/RetroComputing/ZXSpectumAMega/Resized/zxif1_rcx_communication_v4.txt It works ... Here is a snapshot of the "send message" ZX BASIC code: Lines 5020 to line 5040 compose the messages to be sent to the RCX or NXT: ID + databyte. I love that lines 5042 to 5047 don't work: The ZX has these system variables for "measuring" time (there is no millis() or whatever method - there is POKE and PEEK ;)) but that is just not "stable"; 50% of the time it works, and then jumps to almost zero delay, back to what can be expected. A simple FOR/NEXT loop is rock solid, though, timing wise. A quick glance at the machine code representing FOR/NEXT is simply bolted to CPU instruction/clock cycles. I have no idea why POKE and PEEK of the system variables does not work ... but whatever. And here is the reply ZX BASIC code: Here the "short" POKE/PEEK of the timing system variable works. Nice. With this, I believe all the stuff required for an 8-bit stone old 1985 computer controlling LEGO trains is online: ZX Spectrum BASIC code (with Interface 1): https://brickshelf.com/gallery/ThorstenB/RetroComputing/ZXSpectumAMega/Resized/zx_if1_rcx_control_v24.txt Arduino MEGA code: https://brickshelf.com/gallery/ThorstenB/RetroComputing/ZXSpectumAMega/Resized/zxif1_rcx_communication_v4.txt Next up: A little video showing how it works. May take time though. Best Thorsten

No it does not, when a stupid user cannot handle C++ regarding its >most basic< commands, methods and whatever. This now works on all channels; IR, RF, LEGO Tower. Just to make sure: A nested "if, while, if" thing is not good, nor elegant - I am too stupid though to come up with the elegant stuff. The thing is: This works // Listen to Tower, but only when a) data were sent by IF1 (flag_tower_enabled is true) and b) something // actually arrives. The ID+data sent (echo) are flushed above. // The receiving loop is cycled until max_tower_RX_enable_time is reached. // For the RCX this time can be considerably shorter than for the NXT, as the NXT send the reply MSG only after // finishing sending out the PF command(s). This may take a long time (even >3s) when going from +7 to -7 as // the PF power is set +/- 1 at a time if (flag_tower_enabled == true){ // listen to IR/RF/RS232 (RCX reply) on Serial2 // otherwise skip to listen to ZX on Serial1 while (millis() < (tower_RX_enable_time + max_tower_RX_enable_time)){ // record and send out entire Serial2 buffer using HW hand // shake via RTS line control by the ZX if (Serial2.available()){ // wait for data - whatever they are - to arrive ... byte tower_data = Serial2.read(); // store one byte - whatever it is while (digitalRead(pin_RTS_in)==HIGH) { // HW handshake (wait loop): wait while pin_RTS_in is held // HIGH = ZX IF1 accepts no data if (millis() > tower_RX_enable_time + max_tower_RX_enable_time){ // when ZX BASIC program is set to no-handshake-mode, pin_RTS_in // does never go low. Checking for timeout again here, otherwise // this loop becomes a trap. ONLY required for non-handshake // operation selected on the ZX BASIC program Tower_Flush(); // time out has occurred; empty Serial2 input buffer; // RCX may have answered (RCX always replies by default) tower_data = 0; // optional; clear last Serial2 data byte received flag_time_out = true; // only for debug mode - notify time out break; // break out of current while wait loop } } Serial1.write(tower_data); // send out one byte to ZX - whatever it is Delay(10); // give ZX some time. Delay(1) does not work; then not all // bytes of a regular reply are transferred, regardless of HW // handshake control. Mandatory! // Delay(5) good for 2400 baud setting, Delay(10) always stable if (debug == true){ // print some debugging data on the serial monitor Serial.print(tower_data); Serial.print(" "); if (flag_time_out == true){ Serial.print("RTS time out"); flag_time_out = false; } } }//if Serial2.available }//while (millis() < ...) flag_tower_enabled = false; // reset "awaiting data LED" in next main loop via SetTowerLED }//if (flag_tower_enabled == true) More images are here: https://brickshelf.com/cgi-bin/gallery.cgi?f=582185 And more "importantly" (as if anyone cares, dude): I >just< found out, that an ESP32 Espressif board (10 bucks) has - oh crap - 3 UART serial ports. One for the USB, two - for you. Man. The thing is, that that board speaks BLE as well ... is much smaller than the Arduino Mega ... and I used such a board for a stand-alone solution (= no stupid apps on smart devices) for my Croc ... OK. So. What do I do ... ... I'll use this setup with the Mega and my super-duper-shield to do tests on my RCX/PF trains. When that works, switch to an ESP32 board (see, whether this one can cope with the demanding HW protocol of a 1985 computer!) and - well - see what happens! @1974: Ole, can you believe that I am looking forward to trying this crap (above) in September, when we are spending our two-week vacation on Fanoe? Red wine, white wine, Carlsberg Porter (and the locally brewed beers), the North Sea maybe calm, maybe demanding, a restaurant - more pub like - close to the beach, and all this ... slowliness ... Esbjerg not smelling fishy anymore, when the wind comes from the land (well - to be honest: Much, >much< better than seeing all this off-shore oil industry growing there - but - yes - also windmill parts) ... and me, after a what-ever-weather-day, after dinner, after all went to bed ... trying it out? Maybe not All the best Thorsten

Lalala - I started this - lala - thread - lalala - so I can necro-bump it - lalala lala ... Phew. Four months later with apparent total inactivity (yes, semesters are semesters, and our work on the metrology for ASML's EUV scanners is a bit on the time-consuming side as well ...) I finally arrived at this: Front dark blue trans tile: On-board IR RX/TX. Left RS232 port: Connects to LEGO serial tower. Top: Three keys to select IR/RF/RS232 communication. Back (not shown): RS232 to ZX Spectrum Interface 1, USB (power supply and Arduino programming), 9V power supply. As you may recall, I want my ZX Spectrum, a 1985 computer I still own back from the days (and fixed recently, as two of the 4116 memory chips as well as the ULA died earlier), to control some of my trains (either RCX or PF equipped). The point was: The ZX is in control and runs a program that does it all. This is all about using outdated hardware to run outdated trains using an outdated approach. Some call it retro computing. Others a waste of time. For me, it is pure fun, as I feel to be 23 years old again ;) Now, this is only an update on the progress; hardware works. However, there is still something wrong with the software. Not on the ZX, the Sinclair Basic program works nicely; but with the C++ program on the Arduino Mega hiding in that case. No no no, don't get me wrong: It is not the Arduino MEGA doing the control; that one is only there to translate from x baud/no parity/1 stop bit hardwired into the ZX to 2400 baud/1stop/odd parity hardwired into the RCX protocol. Plus channeling the data stream either originating/coming from a) the inbuilt IR, b) a LEGO serial tower, c) from OOK modulated RF resembling the RCX IR protocol. The latter is to reach my PF/RCX driven trains without line of sight, as discussed here on EB several times (I don't want to resurrect these threads ;)) they all have simple RF (433 MHz OOK) receivers. That's inside: and this is the 1980 electronics on the Arduino Mega "shield" (I found the two 74LS02's on a board on an old mass spectrometer in the lab, we have kept for educational purposes. It is not that easy to get the LS series chips anymore - yes I know the HCT series is there - not the point, this is about the 1980s. And yes, I know that a Mega is not from that time, but I need to start somewhere ;)) (Forgive my bad soldering skills. They seem to be still at 1980 level) Experts in the field will notice a 3 port NOR gate realized with diodes (we are talking 2400 bits per second high-speed data here) - a single transistor logic inverter (with a 1nF/10kOhm low pass "filter" on the RF logic line) ... all these components are +30 years old. I tend to stick to old parts in my work shop ... Here are the almost 20 years old RF components: The solid antenna is just to impress - the two 17.something cm long wires are the RF RX/TX modules antennas matching with 433MHz. Summing up: ZX Spectrum/Interface 1 RS232 communication with LEGO IR tower / RCX / NXT (with HiTechnic IR sensor) works bi-directionally: Check Sinclair Basic control program (a couple of hundreds of "code" lines) works: Check Hardware UARTs required to establish serial communication between a X/N/1 and 2400/O/1 protocol required; I could not get any software serial program on any Arduinos to accomplish this; I believe the major problem is that the ZX Interface 1 requires hardware handshake; it has an input buffer of 1 byte (actually 2, but that is even tougher to sync) and then needs to move that byte before listening again. None of the serial software programs available for Arduinos could handle that weird timing. I used an Arduino Mega (T-O-T-A-L overkill, but I guess that is the cheapest (<10€ - clone of course) one - it has four(!) hardware UART ports. Fully configurable. Check. LEGO brick case built - trying to blend that with the ZX Sepctrum case/hardware. Check??? Next: Will simply dump more pics here ;) Arduino RS232 syncing software needs improvement. And when that is done: Demonstrate how the ZX Spectrum (as >brain<) controls trains from the 4.5/12/9/PF era. PUp is only possible with an ESP32 as essential hardware piece - so not that much of retro computing here (but rather simple to implement). Here is to feeling good: Best Thorsten

Sven, whereas I really do share your notion, I'd rather go with a word, a sentence, a phrase, a joke - than seeing a "like" icon. Somewhere on the left or right, as this would not be in the text section of a reply, right? Because the text section of a reply does allow us to solely reply with: , , , , - we don't have to phrase elaborate texts - as of now. What I "like" about not having that button in a "uniting the fans of LEGO" forum is, that, well, German is not the language of the World. Nor is Danish or Bengali. Uniting means to speak one language. Or sort of speaking one: It does not have to be perfect, no way. Not even close. When reading a comment saying "You modell ist very cul", I am much more touched as by a like icon. This person tried. When it comes to elaborate commenting, agree with you. But: Maybe this is the big difference of EB vs ... you name it platform. Just my take. Best Thorsten

This. I totally agree. Also: Maybe even less experienced builders have found a way to construct something others haven't though of. Maybe it is already a very big honor to simply be able to show off our creations in a very educated environment. To me, no reply means: "Noting went terribly wrong". A few replies (let's say two to four) with some text saying something: The ones who are within my universe have taken some time to reply and discuss. Maybe what we have created looks breathtakingly nice to us - including all the time we spent making it perfect - but maybe it is of >zero< interest to others. For us, it was the biggest and toughest effort ever - for others it is a pile of bricks arranged in an uninteresting way - a waste of time. So what? Is it about us or them? It is about us. Inspiration, learning, giving (provided anyone wants something), and most importantly: Having fun making it, posting it, just showing >our< fun. That is all there is for me. It feels good. Best Thorsten

Wow, this is a beautiful terminal. So many details, very nice usage of the ever too much track curves - really nice to look at. And certainly to see trains stop by and take off of rush through!!! OK, whatever station it should capture: This station would for sure better looking if it were looking like yours! Second: What you feel, is the one and most important thing here! When you are looking at yours station (and I guess this will be almost every day - I sure would; this is what I do every day here) and feel good about it: It does not get any better. Third: Oh yes, budget constraints plus size of layout. That is always a chore. However: You really mastered it: I can't see any constraints here - it is just a beautiful building/station! Congratulations on your accomplishment - and thank you for sharing! Best wishes, Thorsten

@Sariel So what's the rush? I mean, these forums will be joined and most of your reasons to leave evaporate, don't they? Yes, there is this "don't criticize" thing here - to hell with them. Just do it, stir it up and enjoy the show. That is my take, at least. As you said: This is a forum for - huh - "adults", whatever that means. +18 maybe? Phew. You know, when I turned 18 I was not even close to becoming an adult. After more than 40 years passed since then - I am still nor sure. Oh well. So long then - and thanks for all the fish. (There are two distinct perspectives on this) All the best - and keep on rocking. Thorsten

No. It is not safe. It is absolutely - with any thinking required - complete. With regard to success: It is more than that. It is breathtaking success. Congratulations. The valve gear alone is one beautiful design. The white rims ... man. Yes. She is complete - as in perfect. Best Thorsten

Wait wait wait - oh my. Is this representing a 1st order perpetuum mobile? Wow. I may have to revise my lectures on Thermodynamics! This could be Big Stuff. Like cold riveted girders with cores of pure Selenium ... Best Thorsten

Well, I don't know about the Cybermaster software, but so far, I'd claim that virtually all LEGO software runs on a Win10 64 bit laptop (I am using) - except for 32 bit USB drivers, e.g. required for the RIS software. OK, needs some tinkering, but works; there seem to be workarounds for every issue (so far); in case of the RIS USB tower and lack of RS232 ports on modern laptops: Use a quality USB to serial converter, make sure the COM port used by converter is < COM9 and all serial LEGO stuff works flawlessly. As far as I can tell. Best Thorsten

Oh crap I simply could not figure out what "Dupont connectors" really are ... I thought any 2.55 mm pitch connectors are just that. So I was talking about the soldering effort ... but it appears as if Dupont connectors are crimped onto the wire, correct, @dr_spock? Is this reference a good one? https://www.instructables.com/Fitting-Dupont-Connectors/ Well, I stand correct then: I am using these: https://www.reichelt.de/de/en/precision-sockets-2-54-mm-1-x-13-straight-mpe-115-1-013-p187277.html?GROUPID=7435&amp;START=0&amp;OFFSET=16&amp;SID=93864a34b28fc8084dd88cc0deedf68bff49fe84015d4da016504&amp;LANGUAGE=EN&amp;&amp;r=1 Reason: Very small footprint. Downside: Needs soldering. Best Thorsten

Des, then I'd really go for that suggestion of @M_slug357! @dr_spock is also using this approach since long, as far as I recall. There were several posts here on EB where he demonstrated that; both for PUp and for PF. I myself have cut at least 10 PUp "cables" to use that approach. As I wasn't aware of the Ali Express stuff at that time (it was early stage PUp time, and as soon as the hubs were becoming cheap on BL, I wanted to connect 9V train motors to these hubs), I purchased a couple of the cheapest PUp device (the lights = LEDs), cut the wires and fitted then via Dupont connectors to the 9V motors. Works beautifully well. So I don't see any reason why that would not work with: (any) motor (cut wire and fit with Dupont male connector) -> Dupont female connector - cable - Dupont male connector -> Dupont female connector soldered onto PUp plug. Lots of soldering though. This way, you can also experiment with different lengths of the extension cable (blue above) or - when not required - connect the motor directly with the modified PUp plug wire. Also, when you use that direction (female side exposed when unconnected) the power source (hub) is not exposing its contacts - otherwise accidentally shortening them is always a "threat" When you take that route, you may want to consider using heat shrink tube to securely isolate the soldering pins of the Dupont connectors. A bit of work, but can only recommend. Worked/works flawlessly for me. Best Thorsten

Never tried, just clipped off the plugs and used them elsewhere. However, no reason to give up here! And your skills are certainly up to the task, I am sure. So more questions to find the solution(s)! Do you have a multi-meter at hand, as @JaBaCaDaBra suggested using for testing? Do you have dumb PUp lights (LEDs) at hand? With regard to the light sensor: Do you get any reasonable readings from the sensor when using your extension cable (i.e. comparable readings from the hub with and without)? Heehee, welcome to the club! I made a custom RS232 cable to connect my XZ Spectrum to my laptop via RS232. In addition to the usual pin mess even with "standardized" DTE to DCE RS232 cables (or better: "who is who" cables, and yes, there are reasons to occasionally revert to that technology ), Sinclair defined their own pin names; they found it more intuitive to call RX TX and TX RX. Same with RTS and CTS. Best Thorsten

Here is to you @shadowmadman Very nice that you are still around!!! Best wishes, Thorsten

And then there is this "mirroring" thing ... (never) "learned" that the hard way with RS232 cables - and I am not talking about null-modem or whatever cables; just 1:1 extension cables. Is pin "1" (or whatever it is called) on the PUp plug wired to the corresponding pin "1" on the PUp socket so that this matches with pin "1" on another PUp plug? The plug pins are a mirror image of the socket pins - when looking onto the solder side. Yeah, sure, I know. You may be well aware of this (of course), it is just that >I< repeatedly screwed up on this. Sigh. Best Thorsten

First, thank you very much @Aanchir, for that post. I'd like to call it a carefully phrased write-up. It is a thoughtful piece, standing out in this entire thread. This. I have no clue, how the mods, and let me be "personal", @Peppermint_M (and all others!!!) are handling all these very diverse issues on this forum without >completely< freaking out. I once complained about a war-scene on EB - and got some (reasonable!) moderator flak for being rude. Well, I "hate" war to the bottom of my heart. And voiced it. Nevertheless, EB is about building whatever in bricks - and when folks like to do that, because they want to, so be it. Point is: I got a personal message from @Peppermint_M explaining it all - and not only "accepted" it, but had the feeling of >having< to bow, tip my hat - with a - much more - than slight nod. This forum is outstanding when it comes to tolerance, free speech, and >required< moderation. I simply don't want to go anywhere else (and I don't) - because I feel a sense of "let us keep this thing together" here. Much more, than anywhere else. On EB, we are discussing war, town, spaceships, trains, LGBTQ+, molds, how-to-run-a-company, pirate ships, cranes, trucks, programs, electronics, mini-figs, color of people, muscle cars, licensed stuff - including SW, help on whatever comes to mind, non-LEGO stuff (the pubs), even competitors, and this list has no end. And 99% of the storage capacity of the EB server/hosting hardware - in a decent way. I love this community. Which is certainly maintained as such a place by the moderators. Here is to you: Sincerely, Thorsten BTW, right now there is a - well - what do you want to call it ... discussion going on in the Technic forum. Phew:

I believe these come with the spike prime stuff - and they really look futuristic for sure! They may be a bit "thin" though - I believe the originals are wider, aren't they? If not, then I would really use these for the futuristic touch! Best Thorsten

Well, look into the semicon world, industry, and market. There are 2 - 5 (currently) nm nodes to be made, with ever-increasing challenges. I believe there are many other industries tackling such challenges. With a myriad of different compounds and/or materials to deal with. ABS + colors is TLGs (production) challenge. Maybe a very low number of other polymers clocks in as well. The other thing is: Everything you said is correct. However, TLG had 60+(!) years to streamline/optimize the production process/channels. 60+ years. Is there any other industry having that much time to refine? Yes, there are. Many. But what they did 60 years ago is not anymore what they are doing today. TLG on the other hand, is still molding ABS pieces. The only thing they have to deal with - when all is reduced to getting the bricks out - hasn't changed for 60+ years. They know how to streamline this process, for sure. It is not the molding, it is what to do with the pieces. Revenue wise that is. TLG is a toy company, but they don't/can't toy around. Best, Thorsten

No, you really are >not< stirring. In contrast. I very much like the polite approach you are always choosing, regardless of energy (in a thread). I also enjoy very much reading through your posts. I am actually learning by doing that. There is (maybe!) a notion here on EB, where mods want to keep it "to the bricks". LGBTQ+ is as close to what TLG produces and announces (bricks, set, and pieces wise), as a LEGO brick is to ABS, considering what TLG is trying to propel recently. Well. Who knows. Let's test/challenge that! Could LGBTQ+, which is, as far as I am concerned, a major development in the evolution of mankind (and all varieties of "man" - as a German, it is tough to choose the right English words here ...), worth a separate thread on EB? There will be so much flak, I am sure, but maybe resorting to a place, where people really care, may be a "focused" option? Just that: An idea. Best Thorsten

I believe you are correct. You can actually see and feel it - take a 1x6 of ABS and a 1x6 of CA and "twist" it ... the result may be not that exciting - both are returning to what they were before, but "differently". ABS has superior stability in sustaining its original geometry - provided no bad things (particularly temperature and of course mechanical attack) happened. CA is never that tolerant, nor does it pop out of the mold with 1 micron accuracy (regardless of the finesse of the mold and the temperature program run for cooling down) or keeping its shape for "ever". ABS does ["ever" defined as: t-today - t-zero; t-zero = beginning of TLG throwing ABS bricks into the world - who knows what may happen to "ever", time goes by so slowly - for those who wait, as Madonna says ;)] Cool that you have such a gem in your hands! All the best, Thorsten