What's new
What's new

Calculating MRR In A Grinding Application

allloutmx

Titanium
Joined
Mar 6, 2013
Location
Rochester, NY
I would like to put some MRR face values to what I am achieving on a grinding application. The part is conical in shape and the machine is much like cylindrical grinder.
I am a decent machine operator but by no means a math wizard...would anyone be capable and willing to share with me how to calculate my material removal rates for something like this?

please and thank you in advance
 
To work a MRR you need to find the volume of material removed in one cut.

Then using your feed rate work out the time of one cut.

This gives you a answer of volume removal by time.

To work out the volume of one cut, work out the cone volume of before and after then subtract.

But beware that as the part diameter decreases the volume removed per cut decreases.

However assuming a grinding application, you probably aren’t removing enough to worry about so I would take that as negligible.
 
Hi alloutmx:
There is an easy way to do this:
Weigh the part before processing.
Load it in the machine and time the part of the cycle where sparks are actually happening to process it from start to finish.
Remove the finished part and weigh it again.

Make your calculation : weight removed /Density = volume removed.
Volume removed /time = MRR.

Unless I fucked up one of the formulas, that should do it.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
If the part is ground at a fixed rpm I think that you can calculate the average surface speed of the part by finding the circumference of the average diameter of the cone (smallest diameter on the cone plus largest diameter on the cone divided by 2 and then multiplied by Pi).

Multiply the circumference by your rpm to get your surface speed per minute.

Multiply surface speed/min by your radial infeed amount per pass (assuming traverse grinding). If you are only infeeding on one side use half your radial infeed amount. This result would be your specific removal rate or Q'. This rate will be inch² or mm² per minute.

Then multiply by your traverse rate to get your MMR. Be sure to convert your your traverse rate or your Q' to the same units before multiplying.

Example:

A conical feature that is about 1" diameter on the small end and 5" diameter on the large end.

Traverse cylindrical grind this feature with a fixed part rpm of 35, .0005" diameter infeed on both sides of traverse, with a traverse rate of 5" per minute.

Average diameter of the cone is 3" ((5+1)/2).

Average circumference of the cone is 9.425" (3*Pi).

Surface speed per minute is 329.875"/minute (9.425*35 rpm).

Q' is .0825 in²/minute (329.875"*.00025(radial infeed amount per pass).

MMR is .4125 in³/minute (.0825*5(traverse rate in in/min).

Disclaimer: I am new to this and very well may be wrong somewhere or everywhere. If I am wrong, I encourage you to educate me.

Sent from my SM-G960U using Tapatalk
 
Why... I do ask that seriously.
I've recently acquired the responsibility to plan processes and program our 4 cnc cylindrical grinders. I have about 1.5 years experience on these machines since 2016. My biggest struggle I feel that I will have is with feeds and speeds. I have a ton of already running and proven jobs to compare parameters to and most of our jobs run the same materials with the same spec wheels. From what I've read maintaining a consistent specific removal rate or Q' should yield very similar results from feature to feature in that material with that wheel. So I've been calculating and documenting Q' and MMR for some of our proven jobs to help me establish a baseline or target for when I need to program new jobs. Is it going to be beneficial for me? I think so, but WTF do I know? LOL. Additionally, the parts we run can be damaged by invisible thermal burn and maintaining a proven Q' should limit the chances of burn before I grind $40000+ of parts that are scrapped due to burn discovered later at mag particle or shot peen.

Sent from my SM-G960U using Tapatalk
 
Here is a sample calculation for a cone. you might use it for the tapering to a point (two sides to a point)

and then figure and add the area of the straight core (the rest of the part/cylinder)for the amount of material.
area of a conical shape - Bing video

Likely best to look at past programs of similar wheel and material and figure the rate of stock removal per minute.

Often plunge grinding with full wheel is the fastest method of stock removal..and then changing to long travel incremental infeed to obtain surface finish and size.

Plunge can be faster because the loading is even to a greater wheel surface.
A tickle dress increment ( every xx take .003 or so) can be employed to just take loading so to increase the rate with losing less wheel.
Pause/rest xx and a tickle dress may help assure non-burning.
 
i had the impression mmr in grinding was calculated as doc x feedrate.

are you monitoring drinding power?
 
MRR at a point in time or across the entire part cycle?
Or across an hour or two where one includes part change, size adjust. dress and other fiddling.
Did the part start as a cone?
In the end what counts is throughput so while this thinking about it good be careful.
A fresh dress every pass will get you high MRR in each cut but terrible part cycle times.
Bob
 
Hi ZEPMaine:
Be careful making too many assumptions about your process from one measure...it's a road to Hell if you're implying outcomes from only the one measure especially if it is indirect.

It's a bit like saying I'm going to grind to a given dimension by knowing the infeed rate and just timing the operation without bothering to actually measure what diameter you got..
Theoretically yeah, you could do that, but only if all the variables are accounted for and if you can time it perfectly and if the infeed rate is perfectly consistent and etc etc...you know what I mean.

SImilarly, in this situation, as Carbide Bob points out you have a bunch of potentially confounding variables that you must identify and control if the MRR is to mean anything.
I understand that in a big grinding house where the process is very well controlled and you're making the same parts over and over, you can make some useful inferences from a measure like this, but with a couple of grinders in a normal job shop where the coolant concentration and purity isn't even controlled (just as an example) you can't infer much of anything from a single measure like this.

So if you want to go there you need to develop all aspects of your grinding processes in tandem if that's not already been done, and you need to track and know the influence of everything relevant, not just the MRR.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
You might consider the fastest feed rates and the slowest the machine is capable of and that is likely the variable.
Part material and hardness, the rigidity of the part or set-up, rigidity of the machine, the wheel-type, surface finish and tolerance needs, the volume of stock needed to be removed, heat build-up.

Here is an MMR calculator..but all is dependant on conditions.
Material Removal Rate in Plunge-Grinder Calculator | Calculate Material Removal Rate in Plunge-Grinder

Some materials can grind as fast as milling or turning..and others are hard to scratch with s grinding wheel.
 
MRR at a point in time or across the entire part cycle?
Or across an hour or two where one includes part change, size adjust. dress and other fiddling.
Did the part start as a cone?
In the end what counts is throughput so while this thinking about it good be careful.
A fresh dress every pass will get you high MRR in each cut but terrible part cycle times.
Bob

i was gonna ask the same questions, but you beat me to the punch :P
 
Hi ZEPMaine:
Be careful making too many assumptions about your process from one measure...it's a road to Hell if you're implying outcomes from only the one measure especially if it is indirect.

It's a bit like saying I'm going to grind to a given dimension by knowing the infeed rate and just timing the operation without bothering to actually measure what diameter you got..
Theoretically yeah, you could do that, but only if all the variables are accounted for and if you can time it perfectly and if the infeed rate is perfectly consistent and etc etc...you know what I mean.

SImilarly, in this situation, as Carbide Bob points out you have a bunch of potentially confounding variables that you must identify and control if the MRR is to mean anything.
I understand that in a big grinding house where the process is very well controlled and you're making the same parts over and over, you can make some useful inferences from a measure like this, but with a couple of grinders in a normal job shop where the coolant concentration and purity isn't even controlled (just as an example) you can't infer much of anything from a single measure like this.

So if you want to go there you need to develop all aspects of your grinding processes in tandem if that's not already been done, and you need to track and know the influence of everything relevant, not just the MRR.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
I am absolutely not using this as a sole measure. I'm using it in my evaluation with as many variables as I can to better understand our current processes.

Sent from my SM-G960U using Tapatalk
 
One reference suggests mild steel @ 1 inch CI per minute with 1O hp
cast iron 2 to 4 CIPM
hard steel 1/2 CIPM
and HSS 1/8 to 3/8 CIPM
Some CPM steels I have ground I think 24" x 1/2" .030 in a half-hour+ ...with a 7HP grinder.
I believe I have run mild steel @ 1/2 CIPM with a 3hp surface grinder doing what is called a yard of sparks hog grinding.

One might run a sample of material on a surface grinder to get a ballpark rate.
 
it seems to me that mrr in units of volume is a performance parameter and not a grinding condition parameter as eg. doc x feedrate (= specific mrr) or grit penetration depth.
 
Unless you're doing creep feed or blanchard grinding, I don't think this is useful at all. Grinding is a finish operation. You're not trying to take off a lot of material, so how much you can take off per unit time is a useless statistic. (Unless you are doing something huge, where it might make more difference ... e.g. blanchard or creep feed. And even then, the usefulness would be a one-time thing comparing different wheels.)

Concentrate on something more important, like how long it takes a guy to go pee or something. Maybe put urinals next to the grinders, would probably save more time.
 
Don't be flippant.

It is important to know how much metal is being taken off and how long it takes.

Firstly to assess how efficiently you are running the process. Things change and doing what you have always done 'cos it works can leave jobs spending much longer in the grinding shop than need be. Which can get real important when the grinders are slammed and work is piling up. Updating methods and materials is lots cheaper than more machines and staff. Conversely you could be pushing too hard on a "usually works" basis and accepting more scrap or poorer tolerance work just to get the jobs through. The tortoise wins sometimes.

The multi-zillion parts a year folk have always had to deal with such things but modern CNC based productivity has pushed the output of smaller shops into the region where such considerations are important. Mighty Big Industries can pay specialist folk to look into such things. Smaller shop guy can't but ignoring things can leave significant money on the table. Its easy to leave 10% behind by not sweating the details.

Secondly to help set the pre-grind tolerances to sensible levels to get the most efficient workflow and utilisation of machines and staff. When the grinding department is slack and everyone else is slammed it may pay to go old school by opening up the pre-grind tolerance and finish requirements accepting that the job spends more time being ground but less time on machining operations freeing machinery up for other work. Conversely if grinding is the bottleneck spending more time machining and exploiting the high finish capability of modern tooling may give best bang for buck.

If you wanna make money you have to sweat the details. To sweat the details you need to know what's going on. If you don't know what's actually going on you can't know which details to sweat. Ultimately its all about how much time a job spends on machines and how much each machine hour costs. Doesn't help that the best balance can be different for different jobs.

Clive
 
Last edited:
MRR at a point in time or across the entire part cycle?
Or across an hour or two where one includes part change, size adjust. dress and other fiddling.
Did the part start as a cone?
In the end what counts is throughput so while this thinking about it good be careful.
A fresh dress every pass will get you high MRR in each cut but terrible part cycle times.
Bob

on this particular substrate i cannot go more than one pass without dressing. at around 45min per pass it has dire consequences when trying to get back into the cut and matching up the surface. so yes- i agree but terrible cycle times is a subjective term here.
 








 
Back
Top