Відео

But I used qtiplot! 😅

I just see that you edited your comment, giving a totally new meaning to it... In more recent parts of this video-series, I tried some calibration with other light sources. Also, soon I will attempt to further improve the system with another calibration method, and I will also evaluate its performance from different aspects. Stay tuned!

Absolutely! It must be the same value.

Hi, and thanks! I don't use it in other videos, because it is large and a bit messy to use it as compared to CPU coolers. CPU coolers are just more compact. If you just use a large bucket of water, what's the point of using a radiator at all? Other than, maybe, avoiding contamination in your cooling circle, probably nothing. You could just directly circulate the water inside the bucket in your whole loop and replace the water if it gets too warm. Regarding the cooler box, I have a whole playlist on my build. You might get some valuable insights and learn from my mistakes. 8x TEC 12703 sounds good! Playlist: ua-cam.com/play/PLaeIi4Gbl1T8SlJY5ODqs2cn5hoXZlIWm.html

​@@CuriousScientist Hi, thanks for the reply. regarding cooler inside the bucket, it's actually the other way around, I don't want to contaminate the water in the bucket. The problem with coolant water is that it goes through pump and inside it mixes lubricant (oil) with water. I know it is a very small but still it does so. And since I'll be dumping the warm water into dirt in nature, i don't want it to have micro particles of oil in it. In terms of preformance I see that this oil cooler provided a few degrees of temp change - was this less/more than the cpu water cooler with which you compare it in the video? Thanks for the playlist! I'll definitely check it out

Hi! Yes, if you want to separate the circles to avoid contamination, it is a good idea. Regarding the performance, I did not notice too much difference and the oil radiator actually had some disadvantages because for example the fans did not cover it entirely. Some viewers suggested to build some skirt around the radiator to improve the airflow through the fins of the radiator. I haven't tested it, but it only makes sense. Also, since it is much cheaper than a CPU cooler radiator, I would not mind a tiny bit less performance.

Hi! I cannot form an opinion on this because I never worked with Python. Sorry. I use C# because I know C#, but I did not pick it because of some particular benefit over Python or anything.

Thanks!

Give it a try then. ;)

@@CuriousScientist oh, you evil 😅

Hi! I explained it in the corresponding long-format video and in the article on my website. Check it.

Hi! Thanks for the support! Yes, those trimpots are not comfortable to use. It can be even dangerous because they are located close to the terminals. If your screwdriver slips while you are trying to adjust them, you can accidentally touch the wrong things. Not really harmful, but might fry the board. If I am looking at the right product, then the difference between this thing and the little USP DC-DC converter you are referring to is that this one can deliver up to 20 A (with active cooling!) while the other one is only up to 4 A. I would say, it is up to you. If you don't need that much current, then the other product might be better due to its compact nature and built-in features. Cheers from Sweden!

@@CuriousScientist thanks for the quick reply. Cheers 👍

@CuriousScientist I was hoping that I could find the different trimpots? Is there a specific name or code?

WXD3590. But pay attention to the resistance! It must be the same as the potentiometer on the board.

@@CuriousScientist thanks for that!😀

Wow, thank you very much! You answered the question in the video within like 5-10 minutes after I published it. 😎I had a feeling that it has something to do with the tube, but due to the lack of my knowledge on these systems, I could not really point out what's wrong. I will look up these tube lens and if it is not a huge investment, I might get one in the future. Thank you once again!

This driver is for stepper motors. That's why you won't find any information about how to connect it with a DC motor. You need a DC motor driver. TB6612FNG is, for example, a good choice if your motor is not too large.

@@CuriousScientist thanks for answering so quickly. I will follow your advise

@@CuriousScientist I have looked for more information about my motor and it is supposed to be a stepper motor. The model is HDM: L2.105.1311/01, SERVO GEAR MOTOR, Subcategory Stepper Motor.

If the motor has only 2 wires, then it is definitely not a stepper motor because a stepper typically has at least 4 wires. 2 for each coils. I tried to search for a datasheet based on the data you provided, but I did not succeed. But if it has 2 wires, it is safe to assume that it is a DC motor. Connect it to a variable power supply, set the current limit to a low value and gradually increase the voltage. If the power supply hits the current limit, increase the limit just a little bit. Observe if the motor is getting hot. If it starts to spin, then it is a DC motor. If it does not spin, then probably something else. If it is a stepper, there must be at least two more wires there somewhere.

@@CuriousScientist I just opened the enclosure and it is a DC motor with 2 cable (+ & -) and it is connected with gears to a 10 turn precision potentiometer like the model 7286R10KL.25 with 3 cable (ccw, cw & s) When I turn the shaft of the motor, the gears connected to the shaft moves the potentiometer at same time

Glad it was helpful!

That's a whole big area itself and it also depends if the circuit is DC or AC. But in DC, as far as I know, it is not practical to connect a capacitor and a motor in series based on the characteristics of a capacitor and an inductor (motor) in a DC circuit.

Go and check the official Arduino Uno page. It says 20 mA. Do they lie about their product on their official website? Yes, the _absolute max rating_ of the ATMEGA328P chip might allow 40 mA, but since that's the absolute max rating, it is safer to not use it at those values. You don't drive your car at 8000 RPM even though the engine can do it. But can it _sustain_ it? It won't always reset, probably depending on which clone is used (lack of protective componentsto cut costs). I also talked about the fuse and I said that sometimes it can be damaged permanently.

@@martincerny9866 the number one thing I have done that has cooked an Arduino has been unprotected high inductance load switching. Be wary of driving small solenoids or motors without some sort of diode protection. The pins HAVE internal diode protection however I have killed a device in this way.

Yup, I mentioned in the corresponding long-format video that motors, for example, need a flyback diode.

@@martincerny9866 I will say this. Don’t question this poster like you know something. He will be like, “do it then”…..and now I have a broken UNO. 🤣🤣🤣

Depends on what kind of load cells. The ones I work with at my job are super expensive.

I guess by series, you mean electrically. A stack is really not needed because you won't be able to pull more heat through the system than the bottom unit can pump. Heatsink on the side is not a good idea as you would create a thermal bridge between the hot and cold sides. A thermal epoxy is typically a worse thermal conductor than a proper thermal paste.

@@CuriousScientist No the water in series one to the next getting hotter with each but thermally in series parallel , electrically parallel. It would be an instant heater you can buy resistive ones for tankless hot water. As for the heat sink it doesn't bridge the hot and cold side it would be on the cold side to bring in more heat to be transferred to the water. Thanks for the info on the thermal paste - you would need to design some sort of fastener to keep a little tension on them. I wonder how much energy you would save vs a resistive heater.

I see. So you basically daisy chain a few of the water cooling block which are cooling the hot side. Let's say in this fashion where larger number means larger hot side temperatures: [1]→[2]→[3]→[4]. Could work, but you cannot use it at too high temperatures! Remember that a Peltier cooler has a delicate inside (check my UA-cam Short video where I opened one up!) and it is soldered. Also, as you are getting warmer and warmer hot side temperatures, the Peltier will be able to pump less and less heat from the cold side, you will rely more and more on the Joule heating. Hence you lose the gain you would expect by also having the Peltier effect. Heat sink on the cold side is of course a good idea for the exact reason you mentioned. But overall, I would not have too high hopes on getting large gains. The temperature differences are not too high here, so you can't expect too big values either.

@@CuriousScientist Remember you want flowing water to come out at temperatures that can't harm people so it doesn't need to exceed 42 degrees Celsius that's the pain threshold. With temperature sensors you could feed your shower with a temperature controlled water feed. You wouldn't want too many of these flowing water heaters in the home but bathroom, kitchen, laundry room might work.

How should I remember if you never mentioned the exact use? 😅Also, so far in my life, I have never seen any taps providing hot water so precisely and I have been to some places. It was always hot water mixed with cold water in the tap... Also, I hope that you are aware of the fact that those instant hot water heaters are literally several kilowatts (2-3 kW) devices!? There's no way you can provide that much with a Peltier. It is definitely not suitable for such task.

It is an HMI, so it needs some input from a human, and then it sends the request/instruction to the microcontroller. Otherwise, the microcontroller can automatically send data to it, and the display can show it.

@@CuriousScientist could the request be a timer in HMI?

AFAIK, the request should come from a microcontroller. It needs to receive a command via the serial port.

Over 9000!

Yes, at the smallest details, they are transistors (of different kinds). But, these chips (not necessarily this one) often contain other components such as resistors and capacitors...etc.

Thank you! It was not easy to squeeze everything into a 1-minute clip, but somehow, I managed. 😅

Thanks! In the future I will announce these things in time so people can prepare if they are interested. Now I am just practicing with everything, and these lives are my exercises. 😄

Did someone share this video there?

@@CuriousScientist Not this video, but your channel

It must be the discussion in Electroboom's thread! 😄

Hi! It goes to the ground, so for example to the negative input pin of the buck converter.

And what would you achieve? The amount of heat that you can remove from the plate is always limited by the Peltier at the bottom. If you stack them, most of the useful cooling power of the bottom cooler is used to get rid of the Joule heating of the rest of the coolers... Your suggestion makes no sense.

Hi! The datasheet of the Pi Pico states 50 mA as the max current from *ALL* GPIO pins in total. This is the maximum IOVDD current in the datasheet of the RP2040. Then, I found this in the datasheet: "Output drive strength can be set to 2mA, 4mA, 8mA or 12mA". So, much lower current per pin and in total as well. This must have something to do with the fact that it is a lower voltage device (1.1 V and 3.3 V), I assume. Do not directly drive a buzzer or a speaker, drive it through a transistor. For a motor, you just need a flyback diode and it should be OK. I briefly mentioned it in the video. But if your drive it with a proper driver, there should not be any problems.

Why would they face each other? The flow is uninterrupted from one side to another. They don't blow against each other. I know what I am doing.

It is explained in my videos, and articles. Please check them.

Then, you most probably exceeded the current rating of the GPIO pins. Lessons learned! 😎

Hi! You're welcome. The link can not be more obvious. It is in the description of this video. Cheers!

Wow, thank you very much for the supportive comment! I'll keep grinding! Cheers!

Thanks! True words! 60/40 solder is definitely better, but for example here in the EU, it is banned. Sure, people can still buy them through different channels, but for example, I can not get it from my favorite electronics supplier.

That is also true!!! I have a roll of cheap solder and the results are absolutely horrendous.

Fantastic remarks, thank you for sharing the knowledge! Yes, this soldering iron came without any regulation whatsoever, and it is probably not the best quality from the get go either. So, it already shows signs of wear and tear. But that's why I picked it for this demo to show that even with this poor tool we can solder reasonably well. Today, we are very spoiled with these nice temperature-regulated irons.

@@CuriousScientist An easy expedient is to connect a traditional iron through a light dimmer.

That's actually a genius idea! Dimming the iron, sound fun and smart.

Watch the video maybe? (@18:17) Or read the article on my website maybe?

@@CuriousScientist no no i mean the optical resolution not pixels

Yeah, it is exactly there. It is written what wavelength range I can cover and with what step size in nanometers... Or what else do you mean?

@@CuriousScientist ou can define a wavelength scale as you whish and you can make many steps or pixels along that path. I mean the true reolution, that is the nearest distance of two vibration level signals that you can see and that makes th difference from a professional 100 000 Euro spectrometer and your DIY toy. Also ou dont tell us the sinal to noise ratio, guess it is verry bad

Probably, the signal to noise ratio is better than your grammar. And don't call it a toy unless you can make better toys from scratch. 🙃 I haven't evaluated the performance of the device in depth to such extent because I lack other equipment that is needed to perform it.

Thanks! Yes, I saw that kind of configuration as well! I am actually thinking of dusting off this project again and build a more solid base for it. I might look into the bobbin holder as well. I also discussed with some viewers about the feeding mechanism and the nozzle. I might revise that part as well. Cheers!

@@CuriousScientist Another option from looking at lots of other projects was getting the nozzle close to the winding seems to matter. So you could have rotate the bobbin as one servo, then an X-Y for the nozzle. But I do like your project!

Thank you! I will research the topic more before implementing new updates.

Thanks! More tutorials will come, both in short and long format. Stay tuned!

Hi! It is not the Arduino but what powers the Arduino. The two ends matter (load and power supply) and what is inbetween these two will suffer. The microcontroller board does not create current out of thin air or something. It regulates what is provided to it. As I said in the video, even though the USB 2.0 only provides 500 mA, you can power the Arduino with a charger (1500 mA) or by a USB 3.0 port (900 mA). Both will knock out the onboard fuse if something on the load side can draw more than 500 mA which is the rating of the fuse. Read my detailed article on this topic and maybe you'll understand it better. Cheers!

Your ideas are all correct, but you don’t seem to understand how it connects Current isn’t what power supply can give(in most cases, there are current regulated power supplys) It’s more what something draws/needs, it’s dependent on resistance, more resistance more current thing need In case when power supply can’t provider enough current, most likely something will fry/overheat and turn off or safety machanism will work, but not only current matters, resistive and inductive loads- lightbulbs, motors will comply to ohms law(there are other types of loads, for example something with integrated buck converter, and it doesn’t comply to ohms law) which means more voltage more current, and more importantly it will always draw current if there’s voltage, and arduino has „internal resistance”(simplified for purpose of explanation) and if you draw current(connect load) through resistor it limits it, but overheats and that’s what fryes it, i hope i managed to explain it understandably

Great remarks! Thanks for the contribution!

@@bartosz7778 In hindsight, indeed, it is the relations between these 3 concepts that I don't understand. What you said is a tad jumbly but I'll try to rephrase it to see if I understood all that well, and you'll correct me where I'm wrong. - So you're telling me that Current is not a physical phenomenon that is PUSHED/SHOVED by the power supply but rather a phenomenon that is PULLED by the Load, ight ? It is like a "reverse pressure". As in, If we make analogy with a fluid like water for example : when You are put into water you will be crushed by the Water Pressure -- which is exatcly the opposite of Electrical Current as in : if YOU WERE THE LOAD and THE WATER PRESSURE IS THE ELECTRICAL CURRENT, You(the Load) will be the one attracting Water Pressure(Electrical Current) instead. - It is practically impossible to dissociate Current from anything/any material that has a Resistance to it. Hence the "internal resistance" of some material. - As long as there is Voltage, no matter how low it is, there will always be a high theoretical "potential" for electric Current (potential here meaning something different from potential=voltage; just a word I used that I find the most describing); Like, a power supply has an undefined quantity of Amps to "give" until there's a Resistor that "demands" _this or that much_ quantity (Amps) has been connected to it. Right ? - Ohm's Law will show us how much Current, with a given Voltage, a Load demands ; and that's why for example 1kOhm Resistance will "call" for 3mA with a 3V power supply, and 5mA with a 5V power supply. ??? - The greatest Load (exteral resistor) will fry the [assembly of] Smaller Loads that are connected to it in series (internal resistance of Arduino here) because the Current that the Bigger Load calls can't be handled by the Smaller Loads as it goes through them.

@@thefanboy3285 1st - kind of, yes but I wouldn't say that something draws/ demands X Amps, it seems wrong, more correctly would be using it with resistance so at x Volts thing draws Y Amps because resistance is Z ohms, it doesn't really suck current, but rather needs it to overcome resistance, (idk if this analogy is right but) its more like thin pipe, you need high pressure to get liquid flowing 2nd only as long as there's voltage ofcourse 3rd(I don't quite understand you here) partially I think you're correct - power supply has voltage, the value of it doesn't really change(transformers can have higher voltage without any load, ex- ~15v for 12v rated, but anyway) yeah as long as there's no load the current doesn't flow, and when load is connected it draws X Amps current, not All what psu have (unless you connect something rated for lower voltage to higher voltage, then all current might flow and it'll fry) 4th yeah, but I'll use "draws" instead of demands it'd be easiest to explain with two resistors one for load(1W, 5 ohms so 1A/5V) other for Arduino(1/4W 25ohms - 20mA), technically you could limit the current for load with resistor so that nothing would fry, but the problem here is wattage, so what resistor could handle and that is how we might compare transistors inside Arduino to big load - lightbulb or motor, you could limit lightbulb current to 20ma it wouldn't light up, barely get warm inside, but you need "strong" resistor also worth mentioning, for you to remember - current goes the path with least resistance, that might be useful and is easy to remember I hope I got all of it right

The tip is typically coated with a thin layer of nickel-chrome plating. If this layer disappears, then your tip is done for. So, in your case, this layer is not there anymore. When you rub it with sandpaper, you remove the oxide layer and when you quickly apply the solder, you create a protective barrier against oxidation. If it works, it works, but it is really a sign to replace that tip! Also, you said "steel wool"... That ruins the protective layer. So if steel wool is really what you are using, then you induced the damage to the tip yourself. We use brass for a reason. It is softer than the plating, so it won't damage it, but it is hard enough to scrape away the dirt from the tip.

Get a chemical soldering tip cleaner like STANOL Tippy or equivalent. It cleans the tip and also tins it. Also using sand paper destroys the tip. So you might want to get a new one.

Super tips! Thanks for the contribution!

Noooooo pls never use sandpaper 😢

Neither steel wool!

Thank you! Yes, excellent remarks regarding the contact time and excess heating. Definitely, a preheated pin with a good thermal connection to the pin or component leg is better because it takes less time to perform the soldering.

It's just a matter of patience and practice. Don't give up!

it's incredibly easy to get sufficient solder joints i don't use flux or wick my joints aren't pretty like this guys but ive been building drones for 4 years now and never had an issue with my connections whatsoever

You might unknowingly use flux because most of the solder has flux in them. Most of the solders are so-called "flux core solders". They have a small channel in the center that has rosin in it. So the solder "applies its own flux" during soldering.

I am glad to hear that you like it!

Thank you so much! I hope people will use this as the "go to video" in the future. 😅I also hope that you will like the future videos too. I compiled a nice table of the topics I want to discuss. I will release them one by one in the near future.

Hi! What is the point of adding a Peltier cooler to the cooling loop? The water cooling loop for the CPU already does what it has to, it keeps the CPU near room temperature. I would rather add 1 more radiator to my loop than adding a Peltier cooler. What is the point??

@@CuriousScientist The idea is to get water temperature below ambient (before CPU) and keep CPU cooler . (A cooler CPU means more power :) Also TEC could be switched on only if more power needed :) ... I agree that when ON mode and water temp below ambient the standard water radiator will heat the water :) which is not OK, but ... if AIM is "keeping water at ambient" using TEC seems easier because radiator will be used in high temp difference mode

Yeah, I get that part, but I am still not getting the whole concept. You will need Peltier coolers (1 won't be enough), you will need to cool the Peltier coolers, they need power supply, power regulator and other accessories (cables, connectors, switches...etc.). Why would you throw several hundred dollars out on this if you can buy a better CPU or a better water cooler? It makes no sense.

@@CuriousScientist My idea is that even with the best cooler when water temperature gets closer to ambient (low temperature difference) the cooling gets more inefficient, BUT if TEC introduced, its cooler difference will be always high and water cooling will be better. No need for power regulators - if TEC wotks in const Voltage efficient mode it will be usefull to get (ambient + 5) to (ambient + 15) degree temp (calculation based on idle CPU temp - TEC profit = ambient + 5 :)... and that will be better than any cooler with any number of radiators ... Seems there is a profit from standard, BUT I may be wrong?

If there's a profit where there's not a single such commercial product on the market? PC users buy all sorts of gadgets... but this cooler somehow never reached the shelves. I wonder why. 🙃

It depends on how you define by "work". It won't be sufficient as an air conditioner.

Thanks! This project can deliver much-much more than that 1 V - 1 mV measurement range. I hope it will be useful for you.

Hi! It would have been great if have checked my cannel before drawing conclusions. I have tested several different Peltier coolers across their whole range and made a very detailed analysis. This video is only about reading and understanding performance charts...

@@CuriousScientist OK, thanks - It seems I missed info. What I ment is that in these series, this TEC in this mode tested seems to be working bad, but more info how to become usable may be would have been a good thing. Did not ment to critisize! TEC series are super - Thank you for them! Pitty there are not many videos in you tube how TEC-s to be used as "chillers" in water CPU cooling ... JUST saw that my comment is not in right video - that explains your reply (I messed video link by wrong copy/paste - comment was about ua-cam.com/video/sfMHF0kdI4A/v-deo.html&ab_channel=CuriousScientist .. sorry for the error)

There are not many videos on the topic you mentioned because they are useless in such applications. Peltier coolers are not meant to be used in such applications. They are used in x-ray detectors (Si-Li drift detectors), CCD detectors, lasers...etc. They are not made for large thermal loads such as a large tank of water with something that spews 100-200 W heat in it...

It doesn't have enough cooling power. Peltiers are useless in those environments.

So make a ac for car that can run on water radiator water pump andd ice 😉

What?

I mean Small radiator plus water pump and Water pump pipes and pc fan or last box for cold water 🤨

I still don't understand what you mean.

Hi! Of course! The fundamental principles are very important. The rest is "just" creativity and diligence. 😄

Thank you!

Hi! Nice catch. When you upload the very first program to this board, like in this case, the COM port can "jump" sometimes. When it was showing COM 26, that was an empty board only with the freshly burned bootloader. Then, I uploaded the code shown on the display via USB. This changed the COM port to COM 27. So that's why, after the upload, the terminal shows COM 27. I just didn't show the part where I changed the port to COM 27.

Hi! Have you done any measurements of the temperature of the bottom plate? It should go below -25°C in order to create the supersaturated alcohol vapour. It sounds like your bottom plate is too warm. Why it is warm is a different story. Have you read my article on my website? I wrote a very detailed description of my system and your system seems to not follow those guidelines. It seems like you are "overdriving" the Peltiers and their waste heat is not dissipated properly. What kind of cooler do you use for the bottom Peltier?

@@CuriousScientist A CPU cooler similar to the one you had. Thermal paste applied between the 12715 and the CPU heat sink, and another layer between the 12715 and the 12712. I do not know the exact temperature of the top cooler attached to the plate, but my temperature gun arrives Monday. Drops of water on it freeze almost instantly though, which means at the very least it is cold. Which sort if implies the liquid condensing on the plate is alcohol and not water, since all the water should just freeze at the bottom.

Let's wait until that thermometer arrives. However, if you bought one of those IR thermometers, it might not be the best choice. A Pt100 or a K-type thermocouple would have been a better choice. In the meantime, you can try decreasing the voltage of the modules. It sounds counterintuitive, but les voltage can lead to lower temperatures due to the less Joule heat being generated.

@@CuriousScientist ok I've tested the temperature, it's only getting to -16C with the 12712 reduced to 3.3V and the 12715 remaining at 12V. Im not sure why it's achieving such poor results considering I got the 12715 and 12712 running at 12V and 6V (which did indeed have worse results) and it worked for the guide I read, but mine doesn't even get close. Not really sure what to do from here

I knew it was the temperature! Well, get a better cooler. Or follow maybe my video instead of the guide you read.