Mist Coolant System Update
Вставка
- Опубліковано 6 вер 2024
- In this episode I am updating my Mist Coolant System. I built a system about a year ago and have been using it but there are several short falls. The main problem is the needle valve delivering the Coolant. The system is pressurized and should deliver a constant flow of liquid to the Spray head but I often have to adjust it.
So I am improving my Current Coolant delivery system by metering the fluid in using a peristaltic pump. I will show you the electronics needed and also how to hook everything up.
If you like a measured drawing and a components list it is free to download here:
www.jbworxstud...
You can also purchase the spray head I made a couple more than I needed.
www.jbworxstud...
Have fun ...
Joerg
i used a Smc02 Stepper Motor Driver for the pump, and mounted it at my operators panel. and also used a relay to control it via the cnc controls. Works really well thank you so much! 😎👍
Yes that is what someone reported as well. It’s an all in one unit- yes? I have one here to test. Someone else reported that he cannot get the rpm low enough. But sounds like you got it going. Great job !
@@JBWorx i goofed around with the settings a bit. you can set a rpm for forward and reverse to. Might have to up the amps on the controller. i used a different pump as well. it's a nice all in one package ( be aware there is a red and a blue one, i have the red one. no need for a separate stepper driver with red) and ofc you need a power supply to give it power
Excellent
You did EXTREMELY well with the product. Having used a FogBuster forever myself I agree with you on the mounting issue right our of the box. Keep it up :)
Thank you PGTDESIGNS, what motivated me to make this change was that I hit start on a brand new part and ran into my vice while I was adjusting the fluid. The machine was going slow and I could’ve just pause it but I was not paying attention.
The needle valve was the main reason for me to go with a pump for ethanol aswell. Soo much better!
Yes, using this a while now and I can say it’s been a great improvement. No fuss anymore with cooling you can concentrate on the machine instead.
Rad design! I bought all the components you showed and realised "there has to be a simpler way". I found ZK-SMC01 Micro-step controller. It has a serial port and other control ports to start and stop it from the cnc controller.
The unit I show is also controlled by either M8 or M7 machine command simply energizing a relay. So you still need an amplifier for that unit ?
I love the setup which you've developed. I have a Tormach PCNC 1100 S3 which I developed a flood and air system for, but even with an enclosure the GPM's and air are too much. I do have a flow control via diversion, and a wash down hose setup; but when I run the Tormach RapidTurn I must run dry. I've explored many designs, but yours appears to be the most practical ($400 for a Fog Buster??) Pump and fluid transfer systems are no issue for me. Retired Lt. FireMedic/ ED-RN (peristaltic pumps for IV's) or 1500GPM+ for Fires, fluid dynamics are the same. I did look at your website, but sadly I found no BOM, drawings or schematics; no problem I did have a couple of yrs. of electronics. Thank you, and best wishes!
Hi John, thx for checking out my webpage. It’s still in development but there is a drawing of the pray head and a component list as a free download.
www.jbworxstudio.com/shop
The use of a peristaltic pump for fluid delivery is a great idea. I've tried two different types of cheap Chinese misters, and have been so frustrated by their inconsistent fluid delivery amount. It seems that just looking at the adjustment screw wrong could cause the fluid delivery amount to change from almost none to a flood level. I was considering purchasing just the genuine 'fog buster' nozzle assembly, but considering the fact I want two and they cost $200USD per, it was something I was having trouble justifying. On top of that, I'm not confident that their fluid output metering is that much better (maybe it is though, I don't know).
I have an old, factory installed, Bijur coolant system on my machine, so I all I needed to add was the nozzles and metering. I saw a recommendation for the ZK-SMC01 controller in another comment, but decided to go with the SMC02, as it has the amplifier built in. I things setup so that the SMC02 receives 24V with an M8 command (flood coolant) - I have mist reserved for pure air-blast. I have the SMC02 configured so that when it receives power and turns on, it will immediately turn on the pump. The SMC02 has a dial to control pump speed and I've mounted the controller so that I can easily see the pump speed and adjust it.
I see what you mean about a small pump size. I ordered a 180ml/min pump as it was slightly cheaper than the 110ml/min and available more quickly through prime. To achieve the correct flow rate I have to spin the pump pretty slowly, which causes pulsations of the fluid output. It would definitely be better to have a smaller output volume and spin the pump at a faster speed.
Fluid flow comes from a pressurized container that also provides the air. I need to do more testing to determine if it's better to siphon fluid from an unpressurized container though. Right now, it takes a while for the system to prime and fluid flow to start. I'm not sure if there's a small leak or if the fluid suction lines are too long.
Well great that you got your system up and running. One trick is to install a smaller inner diameter silicone tubing in the pump. This will result in less volume and you could speed the rotation up a little preventing the pulsation of fluid. Also I would stay with the pressurized fluid container it shortens the priming cycle time that you mentioned. I only need to prime the system once per day and that is it … doable I think. And yes I had the same frustration with my Chinese version and I am super happy that I attempted this project.
That's a nice solution. Silly boy... you asked for our thoughts!
(1) The smaller right-angle fitting for the lubricant is nicer, but could you have used a straight connector beside, and parallel to the air inlet. It would need an angled hole in the aluminium part to meet up with the copper tube. (Alternatively, two holes at 90 degrees, and plug the open end of the cross-hole.)
(2) For the electronics... wait for it... use an Arduino! I know you did not want to use one, but an Arduino Nano in a small case can be a tidy, flexible, and reliable solution. It also opens up opportunities for things like increased mist for the first few seconds of a cut, quick flow rate control, auto-stop of mist when spindle stops etc.
Love your videos, thanks.
These are both very good ideas. The Chinese version of the lubricant and air block has exactly what you describe one hole blocked by a ball. I think it would possibly make it look cleaner or sleeker maybe. An Arduino is the way to go for this Projekt, the initial automatic fast fill of the tubing is a feature worth having. Lucky on this particular setup the pot to full for a couple seconds and then all the way back to low achieves the same thing but that is on my version. For someone else it might differ and to get back to the initial setting might not be as easy. Automatic ON/OFF is now provided by the additional air valve controlled via the M8 command in the G-code. I was afraid using an Arduino would make this solution less “popular” as many not so electronically inclined viewers would simply not be interested in building it. Heck I thought about scrapping the idea once I realized you have to solder some wires on. My goal was to simply plug components together. But yes the Arduino is a nicer solution. My idea list for projects has a router lift on it. That will definitely qualify for an Arduino. I might need help for the programming on that one though.
Excellent idea. Thanks for sharing all the details. 👍👍👍👍
Thank you
With Planet-CNC you can omit the PWM generator. You can regulate it directly via the control controller.
A Thomas pump is sufficient as an air dispenser, I've been using it for years. Initially with the PWM gerator.
I think I can do that in Eding CNC as well. I have several more PWM outputs open however they are meant for additional spindles. I will see about setting up a second one and use that output, you made me curious if this could work. Thx.
@@JBWorx Yes, that's right. If the hardware can output various PWM signals and you can control it via software, it will work.
I clamped the PWM signal directly to the step pin of the motor controller
Excellent! I will give that a shot
I stumbled on your UA-cam channel a few months ago wile making research on how to cut aluminum with my CNC. I found a lot of very useful information thanks to your detailed videos that I find excellent. I have sometimes good results and other time bad results. Your design described here looks very promising for me to cool things down and I went ahead and ordered all the components to try to reproduce your idea.
I started with the electronic and I have everything working with what I think is exactly the components you have except for the peristaltic pump where the pump is driven by a Nema 11 instead of a Nema 17 that you have.
When running the pump, the motor driver gets very hot even with the heat sink, stopping the pump on and off; probably a safety feature when the chip gets too hot. I haven’t put all the electronic in a box yet and everything is fine when I blow some air on the chip and I will probably have to install a fan to cool down the motor driver.
I see you didn’t mention anything about it, but did you get any similar problem with your set-up after using it for a while and if so, how did you resolve the issue?
Yes the chip does get hot but the heat sink takes care of it. Even stuck inside the box it never shuts down. I suspect the amp draw of your motor is too high. So one of my viewers reported that he found a “all in one” driver and amplifier on Amazon. I have it here to test but have not accomplished that yet.
Search for “stepper motor driver controller integrated board forward / reverse” it is sold by Banria store. I will place another response below this trying to add a link.
Stepper Motor Driver Controller... www.amazon.com/dp/B0B1HNS9L2?ref=ppx_pop_mob_ap_share
@@JBWorx I've seen that one on Amazon and ordered it a while ago. Had to return it as it was OK for high speed, but as soon as I was slowing the motor down it would stop. Don't know if it has to do again because the motor is drawing too much amp.
Yes that can happen. Making the motor run smooth on a slow rpm is not always possible with every amplifier. So instead of guessing it would be best to simply place a volt meter set to amps in series. This way you will exactly know that the draw if the motor is.
@@JBWorx Thanks for the advice, I will try the voltmeter as mentioned. The stepper is a Moon's Industries ms11hs5l4100 and it is rated at 1 amp.
Slowly turning the US to metric, good job))
😀😀😎
Great Video!!! Two questions: 1. What it the description for the cable that goes to the peristaltic pump. I would like to source that. Secondly what hold the heatsink on the chip. I have used the heat sink paste, but I thought the heat sink needs to be held in place somehow. Thanks... Richard
Hi Richard, the cable is an ordinary stepper motor connector cable - mine came with the board or pump can’t remember. If you have to buy it separately just look for the NEMA size of the motor so let’s say NEMA 14 and search for connector wire. These have usually a JST connector on the end the number gives the pitch between the individual connectors in the plug. So you can check that as well on the pump itself.
The heat sink has a tape on the back and that is what I used. But you could upgrade to thermal paste and another way to hold it in place like 2 stand offs and a bridge over it or a small wire tie. Another idea is to remove some of the tape and use both in combination. Hope this helps
Nice to see, thanks. Just to check I'm reading this right: did you go the stepper pump route, because you didn't want a needle valve, and you can control the flow precisely with the pump? So did you use compressed air previously, to push coolant out of the tank?
Yes, all correct. I did use compressed air on the coolant tank to push the fluid out. The needle valve gave me always trouble in the beginning of the day and I had to make adjustments frequently. Now I turn the knob to full for a a couple of seconds for the first job to prime and lucky for me the knob all the way down is just the right amount of coolant. I like this so much better. And also adding the automatic on off via the 24V valve was definitely worth it. The program now turns it on at the beginning of a job but more important off at the end. If you happen not to be there you don’t come back to a coolant puddle. The switch enables you to turn the coolant off and just have an air blast.
Love that surface finish！
Thank you Kurt, and the funny thing is I have no idea what causes it. It looks a bit mesmerizing almost like a checker pattern.
At some angle with the lighting，it really looks like it was scraped by using the Swiss pull scraper.@@JBWorx
You know what, I never thought about that but that is exactly how it looks like.
Nice project! However; I would like to see the whole setup? Where is the pump and the reservoir located? What is the approx length of the tubes between pump and nozzle an between reservoir and pump?
I think the mounting will be unique to everyone’s machine. On my machine the pump sits in the front where I can easily reach the speed dial and the coolant canister is on the right side next to the control cabinet. The tubes are about 2-3m long. A lot of people contacted me for information on this project and I know that several machines are rather big with tubes of 5m and longer. The only thing g that will change is the time for priming the system.
@@JBWorx Okay, nice to know that tube length does not really affect the workings. Thanks!
👍👍
Nice mister mister! A low flow pump is a good idea! I'd use a 3d printer ramps board that I've bot laying around, with an arduino 2560. I can control it with a laptop using gcode at any speed and length of time.
Still, I'd like to control everytging from the controller and gcode itself though. Maybe a normal solenoid to cut power to the pump controller that is controller by the same pin as the sokenoid valvue?
Yes that is an excellent solution. Actually that is exactly how it is wired. The pump power comes from the relay that tunes the air on. The additional on off switch for the pump is for “air only”. Using an Arduino is a good solution if you have the skills to program it.
@@JBWorx Sounds good! Getting Marlin on an arduino isn't too bad, you don't need programmer skills for that.
Yes I made a couple of arduino projects for my first drones as systems at that time were not available. Many people however will not attempt to build something with an Arduino in it. Great that you are able to do it so many
More options.
By the way i dont understand why you need air presure to the tank, That small pumpe should able to suck fluid up from below 5-10 meter. so the tank could be maby a coca cola bottle. Also i think that you could put the small 3mm tube into the Lock line, think it would be easy , and look a lot better.
The pump is just for metering the liquid in and it primes way faster with the pressurized tank. Easy to do and worth it. The cheap Chinese fogbuster knock of has the fluid hose inside the lock-line. Different concept as the mixing happens at the tip here the mixing happens in the Aluminium block and liquid and air is pushed into the brass tube. Yes, you can use a beverage plastic bottle but there are much sturdier versions than a coke bottle. Once you have that in mind and start watching these bottles you will notice that some are incredibly stout and well suited for this application. I did however not mention it as I am afraid someone will put full compressor pressure on the bottle and it will burst. But yes it will work just as functional and much cheaper. As always good suggestions from you, thank you Lars.
BTW, I liked and subscribed.
Thank you John
Great idea, but strange you did not use a standard stm32 mcu for 3-8 axis/motor boards, like Makerbase or Bigtreetech, they come with 5,12,24,35,48v stepperdrivers out of the box and are easily programmable and can integrate into most cnc controllers via spi,uart,i2c etc. they cost about 30-60$ . Anyway , really like your videos.
Thank you for the suggestion. Maybe a bit $$$ there is a finished pump that I M using now from Kmoer that is fully programmable out of the box for about $120
Hi, awesome channel for beginners like me! thanks!
though I don't quite follow how is the tank connected to the rest, the coolant is still pressurised when using the peristaltic pump?
Thank you, yes the task of the pump is not to build pressure against the air pressure but just to meter in the exact amount needed. And that repeatedly. Works great give it a try..
Great video
Thanks!
I wish you to sell all of them as ready-to-use :)
You mean everything pump etc in one kit ?
yeap :) @@JBWorx
even the white box :D @@JBWorx
Ok if you like you can reach out on my website there is ready to go pump and I have a new type of canister that I am using. It’s not cheap about 285$ all complete.
I like the video but please, edit the sound level as i cannot hear anything, you are super quiet compared to all other videos on YT
Hmm interesting. Thx for letting me know
Heya Jorg, setting out to order parts for one of these. I am pretty lucky that I can tap into the extensive, but local used market in China but I think you have found the flow rate of yours is way higher than it needs to be, right? I am guessing you don't even use it at 1/10 of its rated output...? Maybe more like 1/50th?
3ml/min that's about a cup an hour. I'd be surprised if you use that much, even right?
If you feel like it, could you run it someday for a min or two and check "exactly" how much you spat out;)?
Yes you are correct it is very little fluid. About 1.5 to 3 ml depending what coolant I use.
@@JBWorx Perfect:). I am about to buy two at an incredibly sweet deal. For a stepper peristaltic this is about as easy as they come since it has a a board built in and 15 speed settings via a turnable switch on the board or an external pot. Basically just have to feed it 24V. Can also hook it up to the controller later on if I feel like it as it can take speed commands via analog V or mA and 1-2 more ways:).
It has a flow of 0.02ml/min @ 1rpm - 21ml/min @100rpm. So that sounds about perfect:)
They have one with a bit more flow rate, too which can still go low enough depending on what tube you put in. Maybe I will get one of each:)
Yes that sounds like a good unit, the delivery amount looks good. Most pumps put out too much fluid and can not be regulated low enough.
@@JBWorx Yes, this deal is a feel good, from new, they are about seven times more and even moreso on Aliexpress.
I had initially planned on designing my own CNC to be made in granite, but at the same time I have been keeping an eye on used machines. And I sadly just missed out on a amazing deal on a old Taiwanese made machine base/frame. Less than USD 600 for a machine that was designed to make molds so had plenty of stiffness and precision whilst still not being huge:(.
I have already bought a spindle and some nice second hand Delta servos so that machine would have been perfect. This pump deal doesn't make up for that loss, but still nice, haha.
Anyways, I realized a while back that designing my own machine without ever having cut anything is also a bit frustrating, so I think now, I will go with a cheaper second hand machine just to get started. Might pull the trigger on one today. And they say, once you have one machine you will want another... So, this will be a great learner;).
I think you are one the right track. You can learn with that one and then make your own custom parts to improve it. Very rewarding.
How many psi are you running? I either can't get enough airflow to the nozzle or the pressure causes the coolant to push past the pump. Thanks
20 psi that is all that is needed. Ok if that happens it’s because you have not pressurized the reservoir. So my system puts the air pressure into the fluid reservoir and then 2 lines come out one is air to the nozzle and the other is fluid to the line. Hope this helps.
Any reason not to add the fluid to the flow in the aluminum mounting block vs the bottom block? I also hate those little needle valves. Nice job!!
Yes, I think the regulation becomes more erratic (or better said Inconsistent) You can see this already with the short brass extension. And thank you !
If you are using a peristaltic pump to deliver the coolant, why does the reservoir need to be pressurised?
The main task of the pump is to meter the coolant in not to suck it out of the reservoir. I think it works also with the pump alone however it will take quite a while to get the coolant to the nozzle on startup.
so good,…
Thx 🙏
I am installing the air and fluid lines in my DIY CNC build. I have two 6mm lines installed, one for air and one for coolant. As I look on the web it looks like the coolant line is 4mm mostly. Is this critical for the system to work? I will be drawing fluid about 25 ft from the reservoir. Thanks!
I don’t think it will make a difference other than the priming will take longer because of the higher volume of liquid needed to fill the larger tube. Good luck on your build
Thanks for taking the time to reply. I have really learned from your channel!@@JBWorx
👍👍
Any more available. Sir
My plan is to make one more batch next week (April 2024).
Could you please share tips on how you get those beautiful surface finishes?
I think it changed in the moment I added the second spindle mount. Also high rpm like 18-20k and a finishing pass just taking of 0.03mm makes a big difference. Have fun Jörg
@@JBWorx that worked great, I slowed the federate to half on the last 0.03mm and it made a big difference. Just need to work out how to slow it within fusion 360, might be easiest to setup 2 paths.
@@MrBrettStar nice that this worked for you. Fusion let’s you set a separate feed rate for the finish pass. I am sure you will find it nex time you are in the CAM Module.
Couldn't this also be achieved with a paristaltic pump that runs off a DC motor, then drive that with a small adjustable power supply? It seems like it would avoid most of the control boards needed for the stepper.
Yes, however there needs to be a gear reduction as well to get that low rpm and required flow. The pump turns really slow, perhaps with a PWM Control for the DC Motor it can work. There are pumps available with such motors, I don’t like them because it’s a brushed sleeved motor that will just not last all that long. I think a stepper is a better solution. There is also a all in one control board now available. Maybe an update to this video would be good.
@@JBWorx Wow, fast reply :) Yes, agree about a weenie brushed sleeved motor not lasting, and needing to be geared. Was your goal to have the coolant flow rate adjustable via G code? If it is, is that ever used, or is a simple knob more intuitive about how much is enough?
I use just simply a knob to control the volume. I also found that I need a very small amount so really a very slow speed from the pump. @@kurtbilinski1723
@@JBWorx An all-in-one driver might be this one on Amazon, called "Stepper Motor Driver Controller Integrated Board Forward/Reverse Pulse Speed Angle Control Module PLC Serial Communication for NEMA 17 23 Motor" Perhaps it's the one you mentioned?
At first I thought no this is not going to work as it is missing the amplifier. Well looking at the wire diagram and reading the text … heck yes this might work. It can handle 4A that is plenty to run the pump. Thank you for pointing to this it’s the lowest cost solution I have seen as of today.
Even Better.....Stepper motor control with arduino - Attiny84 & A4988 .......no progamming nedded
No programming ? I have to check that out !!!
I cant find any drawings on the webiste for the fog buster or fixture plate
It’s in the shop section.
www.jbworxstudio.com/shop
@@JBWorx interesting on my phone i see a download button but not in my computer. Thanks. As a matter of interest what was the total machining time on the fixture plate also what software did you use. Fusion360 ? If so would you consider posting the files and tool pathing. That was a great video
No problem! My cuts are usually rather conservative so the total time was rather long … I think 4h or so. I use Fusion 360 for design and CAM. If there is interest I can make them available as a download as well. Let me get the low profile vice finished and I will revisit the fixture plate.
@@JBWorx That would be excellent. The files for the mod vices would be fantastic as well. I thought it was interesting that you made the holes at 10mm which is a bit larger than I have seen on other fixture plates. Was that for ease of the threading? I just stumbled across your channel todaay and have been watching your videos. Great channel and explanations.
Thank you Mark! Glad you found me 😀. 10mm for locating on the machine plate and 12mm on the fixture plate with M10 threads. Mainly for strength as it is an aluminum plate.
Nice idea, what coolant do you use? Thanks
In this video it’s alcohol. That’s what I mainly use as of late. You want to look into the concerns using it before you make a change. However I also use Trim Microsol 585xt. It is clear and works well for me.
Seems like a simple brushed DC pump motor would be far less expensive than a stepper. The motor system seems overly complex, I would think a pump doesn't need that level of precision to require a stepper...
The problem with a regular DC motor on a peristaltic pump is that they do not deliver the torque at very low speeds. So some are available with a gear dive. However they are not for long term use and this is why you will encounter stepper motors on quality peristaltic pump set up’s.
@@JBWorx Why is a peristaltic pump required here though? I was suggesting just using a preassembled pump unit. I'm going to need one of these myself soon, so I'm very interested in this, but I didn't see an explanation of some of your decisions, some background on the requirements and why a simpler solution won't work would be helpful.
Yes you can certainly use a finished assembled pump as well. The key is that it will need to deliver a very very low amount of liquid per minute. These are often metering pumps that can do that. There are mainly two designs used in laboratory and medical applications namely the peristaltic pump and a piston pump (LMI metering pumps are all piston design). Unless you get them used they are often rather expensive so I wanted to design and build a system that everyone can easily make and it is reasonable cheap. I have used it for several Projekts now and really like it. It is very consistent and does not leak after it is turned off. Overall a win for me.
@@JBWorx Okay, totally understandable. How much air does this system use? My spindle cooling compressor is small (but very quiet) but these spray systems seem to use more air than one would expect.
The key is to keep the air pressure down. In the beginning I used 2 bar (30psi) and that was really not necessary. Now it’s below 0.5bar (7psi( for most set ups. A air compressor with 2.2 cfm will keep up with the requirement in my experience. That is a small compressor.
No HOW it works. No details. Just a video of how you did it, but not even a single word of explanation of what you are doing. I won't give this a dislike, but nor will I give it a like. However it is definitely not what I want to spend my time watching.
I think it helps if you had watched the first one I built, but none the less thank you for your feedback.
36820-MS. Multifunctional Stepper Motor Drive-Control Board 5USD
Didn't know that one, seems a very nice module for this purpose! thanks!
@winandd8649 thank you