The grinding of carbide tools is actually an extrusion process. Little chunks of the tungsten carbide and the cobalt binder are pulled from the carbide tool tip. These chunks form sludge in the bottom of the coolant tank and the cobalt slowly dissolves or leaches into the coolant.
If the machine coolant is recirculated these fine particles are pumped up to the spray head and sprayed onto the work area. Some of the particles are thrown into the air where they can be inhaled. Other particles are reground into even finer particles. These recirculated chunks will cause rougher grinds and will increase the chance of burning during grinding.
Filtering out the chunks improves grinding coolant several ways. Shops report that filtering doubles coolant life. This varies according to which shops and how they measure performance, but double coolant life seems to be a minimum performance improvement. Cleaner coolant gives better, smoother, faster grinds and helps prevent any possibility of burning.
Filtering grinding coolant removes chunks of tungsten carbide, cobalt and grinding wheels. A dirty sump can have as many as eighty million particles in a cubic centimeter. The coolant pumped out of this sump to the spray head can still have as many twenty two thousand particles per cubic centimeter. Filtering this coolant can remove as much as ninety nine percent of all particulate matter.
Filtering grinding coolant prevent bacteria growth. Bacteria grow in dirty sumps for two reasons. The sludge is an excellent place for it to grow. Tramp oils seal off the coolant and keep air out of it. Filtering coolant removes sludge and tramp oils and prevents bacteria growth. A filter system prevents bacteria growth by removing the sludge where bacteria grow. Filtering also removes tramp oils, which create an anaerobic environment.
Filtering can change the physical characteristics of the coolant. What changes and how much it changes depends on the filter and the coolant. Filtering removes tramp oils from coolants and it can also remove some lubricity from some coolants. In most cases the user decides that the difference is minimal and is easily made up by additives.
Particle counts (parts per cubic centimeter)
Size (Microns) |
In Sump – Unfiltered |
Filtered |
Unused |
0 |
17,209 |
0 |
|
1 |
140,317 |
25,575 |
11 |
2 |
14,382,515 |
21,432 |
1,049 |
3 |
15,364,737 |
9,720 |
1,935 |
4 |
19,644,411 |
4,223 |
3,367 |
5 |
13,751,087 |
2,550 |
3,618 |
6 |
9,120,620 |
1,673 |
1,181 |
7 |
1,894,282 |
558 |
372 |
8 |
631,427 |
239 |
142 |
9 |
420,952 |
80 |
55 |
10 |
280,634 |
478 |
66 |
11 |
0 |
319 |
22 |
12 |
140,317 |
0 |
0 |
13 |
70,159 |
159 |
22 |
14 |
70,159 |
0 |
0 |
15 |
140,317 |
80 |
11 |
16 |
70,159 |
0 |
29 |
17 |
65,774 |
32 |
5 |
18 |
85,506 |
112 |
0 |
19 |
26,309 |
80 |
0 |
20 |
0 |
48 |
0 |
Totals |
76,299,682 |
84,583 |
11,885 |
Notes:
1. In something as dirty as coolants the laser counter defaults to certain numbers that are accurate relative to one another, but that may not be exact counts to the individual particle.
2. The large number of sub-micron particles in filtered coolant reflects the fact that this is a one-micron filter system. The tightest specifications were typically 0.001” so one micron filtering was felt to be fine enough.
Metal content for disposal
mg/L |
New |
Dirty |
Filtered |
Cobalt |
0.138 |
3,210 |
299 |
Arsenic |
|||
Barium |
0.702 |
0.421 |
|
Cadmium |
0.06 |
0.038 |
|
Chromium |
0.024 |
||
Lead |
0.09 |
||
Mercury |
|||
Selenium |
|||
Silver |
All prices are in USD Copyright 2025 Carbide Processors. Sitemap | Website by oBundle
e, but the one adaptor allows me to use annual cutters in an older mag drill, several different drill presses and milling machines without having to own separate adaptors with shanks that match several morse taper sizes and R8 tooling. Read LessRead less about This is a well machined adaptor that
e Ridgid cast iron table and the Incra fence per the Incra directions. It all works really well. The set up took a little while but it’s sweet! I have done probably 50 cuts and each time I measure with a caliper and it’s dead on front to back on the board I am cutting. I think I am within .004 between blade and fence discrepancies. So for a Ridgid R4512 I’ll take it. Now I have a $1000 package that can hang in terms of accuracy with a nicer cabinet saw. Read LessRead less about So I added the TS-LS 32” to my
, the deal was too good to ignore and figured I would challenge any irregularities with the credit card company later. I also located a coupon code that made the price unbeatable. It must be stated, the expected delivery time on this particular unit was 6 weeks out, adding to my concern. This is the same time frame advised on the Incra web site as well. At probably 5 weeks, I was thrilled to receive delivery notice emails. The large unit came in pristine condition and packaged extremely well. I WILL NOT HESITATE TO BUY FROM CARBIDE PROCESSORS IN THE FUTURE. Thank you for the fantastic deal! Read LessRead less about I had been looking for several weeks