Automatic Coolant Top-Up for CNC Machines

kunalv

Member
Join Date
Sep 2019
Location
India
Posts
28
Hello guys,

I am designing a liquid dispensing system for CNC machines that will top-up a mixed solution of coolant & water at the required concentration.

Present concentration in the machine sump will almost always be richer than required, so I need to 'dose' the makeup liquid at a concentration of 2-3% to balance the concentration in the sump.

Water will be supplied under gravity through the plant's service line while the coolant shall be pumped from a barrel. I don't know if infeed of coolant can be automated as well since I've been trying to work on this for a while now and have not been able to come up with any viable solutions.

I have done received some data from clients wherein I have concluded the following :-

1. Daily top-up is done once in the morning (start of the first shift).
2. Top-up volumes range from 20-50 Litres.

So a daily top-up of 20-50 Litres @ concentration of 2-3% is what is required. Water flowing under gravity has to be mixed in-line with the coolant and the mixed solution supplied to the tank through a hose. I cannot have an intermediate mixing tank as there is severe crux for space, so the mixing has to be done in the pipes.

Queries :-

1. I have thought about using an Injection Quill to inject the coolant concentrate into the water pipe, but i am not sure if it will help.

2. Also, a dosing pump will dose in pulses, while the water flow is continuous. Can I get the right mixing efficiency in such a case?

3. Will Hall-Effect Turbine flow sensors be accurate enough for such an application to sense the flow-rate?

4. Are there any compact motorized valves I can use to throttle the flow of water?

5. I am planning to use a Capacitive Level Transmitter inserted into the tank to know the volume of liquid present and how much will be required.

6. The present concentration in the tank shall be fed into the HMI by the user. The PLC will then actuate the process with these parameters viz. makeup volume and concentration.

If someone can guide me as to how I can incorporate all this into a workable solution, I'd be forever grateful.

Thanks,

KV
 
So you are basically trying to automate a refractometer measurement.
Like you said the concentration of the mixture in the coolant tank will always become higher because the water will eventually evaporate over time.

What is the makeup liquid you are speaking about?
I would think as the water evaporates you need to add more water, as you do the "pumps" in the tank should be able to mix the solution on its own, since it is circulating from the front of machine to the back if there is a bedwash feature, mostly there should be at least 2-3 pumps in there for different coolant needs?
 
So you are basically trying to automate a refractometer measurement.
Like you said the concentration of the mixture in the coolant tank will always become higher because the water will eventually evaporate over time.

What is the makeup liquid you are speaking about?
I would think as the water evaporates you need to add more water, as you do the "pumps" in the tank should be able to mix the solution on its own, since it is circulating from the front of machine to the back if there is a bedwash feature, mostly there should be at least 2-3 pumps in there for different coolant needs?

Ideally I would have loved to automate the refractometer measurement using an Online Refractometer, but the costs are prohibitive and the need for constant cleaning of the prism surface.

So what I am trying to do here is the User checks the concentration in the machine tank using his handheld refractometer and feeds that data into the PLC of my system which then goes ahead and performs the necessary make-up action.

The make-up liquid is the same coolant in the tank, but at a lesser concentration to compensate for the increase in concentration in the machine tank. Yes, more water will be required. As mentioned in the OP, around 20-50 Litres of liquid makeup is required @2-3 % concentration which translates to around 400ml-1500ml of coolant, and the rest is water.

Yes, the mixing should occur in the tank itself due to the constant turbulence, but coolant manufacturers & even machine manufacturers like Haas strongly suggest adding only a properly mixed coolant emulsion into the tank.

I'm not quite sure I got your last point. Please elaborate.
 
You can use a static mixer to have a good mixing and motorized valve ( 4-20mA) with flometers to adjust the flow of both coolant and water.
 
I understand.

My last comment meant nothing really, just asking if their were multiple pumps in the tank to produce some turbulance like you said.

Well i can tell you that some customers like to add water and coolant while the machine is running, and they dont account for a percentage of it being within the machine and not the tank, so when they go to change shifts and Estop the machine the tank overflows. Seen that quite a few times.
 
There are a few things I don't fully understand. You talk about the concentration increasing over time (be it evaporation or what ever). Currently you only top up once a day ?. you want to automate it.
Surely over the day the concentration increases then and you say 2-3% this is already a variation in dosing & over time.
I don't believe that even dosing at a set concentration say twice a day would be that far out if you get the right mix.
As this is not a regular top up you could just dose on a level & after say a couple of doses flag a manual check required & adjust the concentration.
I do not know your process but an example is two flow meters set to dose in correct levels (could set dosing pumps so that the dose speed is 97 and 3 ltrs per minute or what ever time and ratio that is standard), adjustable intervention i.e. operator to be able to vary +- X values to compensate on flag up of manual check of concentration. Even a push button to add a little more if concentration was below levels, so many options.

EDIT: Just remembered something from years ago. We used a number of automatic CIP systems where the concentration of caustic was controlled by Conductivity meters, however one process that was introduced on a plant was with a different solution, there appeared to be no (or cheap method) of testing the concentration in-line, however it was found by experiment that the conductivity changed in a reasonably linear way and by experiment we realised that this was accurate enough to use an in-line conductivity probe. This proved to be very good as the temperature was very constant. Just a thought.
 
Last edited:

Similar Topics

Hello everyone, I'm having trouble solving this one and was hoping someone could help. I have a AB CompactLogix L16ER PLC connected to a AB...
Replies
2
Views
551
Hello everyone, I was hoping someone with a lot of experience with Automatic Device Configuration would chime in with any of your thoughts on this...
Replies
6
Views
1,793
Hello. At my job I get from clients lists of their inputs and outputs (hundreds of them) and what they want the program to do. So I have to type...
Replies
2
Views
1,313
For some reason the PLC (TM221CE16T) saves all Memory Bits and Memory Words automatically so after a cold restart it restores not only %MW0 to...
Replies
18
Views
4,713
Hi, I get an error message "version on your device is not compatible with WinCC Flexible" I know i have to update the OS, and i found a document...
Replies
34
Views
8,346
Back
Top Bottom