Simple device to measure ink viscosity

[Grandad35]

This post will be of interest only to those interested in the technical details of measuring ink viscosity - don't say that you weren't warned.

There are often references to the ink's "viscosity" being off, and causing printing problems. Viscosity is the resistance of a fluid to flow - a fluid with a high viscosity does not want to flow as easily as a fluid with a low viscosity. To give a few reference points, these two links give the viscosity of water as 0.89 centipoise (http://dwb.unl.edu/Chemistry/MicroScale/MScale05.html) and 1 centistoke (http://www.sizes.com/units/centistokes.htm). This is interesting, since water has a density of 1.000 (grams/CC), and the centipoise and centistokes values should be the same for a fluid with a density of 1. For the purposes of this discussion, let's just use a value of 1 centipoise/centistoke for the viscosity of water.

For years, the "Zahn Cup" viscometer (viscosity meter) has been used to measure the viscosity of low viscosity fluids (http://www.geneq.com/catalog/en/zahn_cup.html). The appeal of this device is its simplicity. It consists of a cup with a calibrated hole in the bottom - you just dip it in the fluid that you want to measure and time how long it takes for the cup to empty.

This link gives the ideal viscosity range for inkjet inks to be in the range of 2 to 8 centipoise, so we need a simple device that will measure fluids with viscosities slightly higher than water to measure the viscosity of inkjet inks. (http://www.imaging.org/resources/leinkjet/part4.cfm).

I tried using one of my 30 CC syringes with a 1.5" long needle as a "nifty-cup", and it worked almost perfectly. I poured about 10 CCs of water into the top of the syringe, waited for the water level to drop to 7 CCs, then timed how long it took for the water level to drop to 2 CCs. In my case, this took about 15 seconds. I then repeated the test with various colors of my bulk inks, and the times were all very close to 35 seconds. Therefore, my ink viscosities are all close to 2.3 centipoise (35/15 x 1.0 - my nifty-cup took 2.3 times longer to empty when using ink than it did with water, so the ink viscosity is 2.3 times the viscosity of water).

It is much safer to measure your ink's viscosity than to blindly add water or glycerin to an ink that may already be within spec. There are a number of other properties (e.g. surface tension) that can also affect the ink's performance, and you may actually make your printing problem worse.

Notes:
1. Don't worry if your nifty-cup or procedure is different than mine since we are using a ratio of the times for water and ink. If your rig takes longer than 30 seconds to empty with water, you might want to use a shorter needle or a larger bore needle just to keep the test times reasonable.
2. You might find a situation where the water/ink doesn't flow due to an air-lock - just flick the needle with your finger to start the flow.
3. The ink will form much smaller drops than the water - don't worry about this. The small drop size indicates that the ink has a lower surface tension than water. (http://hyperphysics.phy-astr.gsu.edu/hbase/surten.html)

 

[Grandad35]

This is a companion test to the ink viscosity test that will allow you to measure an approximation of your ink's surface tension.

This link gives the "proper" way to measure surface tension using this technique, listed under "Pendant drop shape Analysis". (http://www.ksvinc.com/surface_tension1.htm) Since we are only looking for relative values for different inks, we will take a few liberties that make our measurements much easier at the expense of absolute accuracy.

We need the same type of syringe that we used in the viscosity measurements, but with the needle carefully ground to a square end with no burrs or distortion of the needle (it must be round, so don't just cut it with pliers - use a grinder and go at it slowly). For this test, the largest diameter needle that you have is preferred. You will also need a short length of soda straw (approx 1.5" long) that is held vertically and sealed on its bottom end - I did this by sticking the straw into a piece of modeling clay. The straw is the container that we will use to measure the total volume of fluid.

The basic concept behind this test is that a drop of fluid that forms at the end of a needle will grow until the weight of the drop exceeds the fluid's surface tension, at which point the drop will fall off. By counting the number of drops of water and ink that it takes to fill the soda straw to a certain level (say 1"), we can approximately determine the surface tension of the ink, given that water has a surface tension of 72 dynes/cm (http://www.ksvinc.com/surface_tension1.htm).

Test procedure:
1. Fill the syringe with a few CCs of water, then hold the syringe vertically and slowly push the plunger so that drops are slowly formed and drop off of the needle at a rate no faster than 1 drop/second. Once you have developed the feel for this, hold the end of the needle above the open end of the straw, but DO NOT TOUCH THE STRAW at any time to avoid affecting the drop size. Simply count the number of drops that it takes to fill the straw to the 1" mark (or whatever height that you want to use).
2. Repeat with the ink that you want to test. The ink drops should be smaller than the water drops.
3. The surface tension of the ink will be approximately 72 x (# of water drops) / (# of ink drops) x (density of the fluid {1.0 for most inks}). In a "down and dirty" test to quickly test the concept, my ink had a surface tension of about 35 dynes/cm.

Yes, I know that this test ignores such things as density differences, drop shape effects, etc., but we are mainly looking for relative values between different inks, not for absolute accuracy.

4/25/05 Edit:
I took the time to test all 8 of my ink colors, and measured surface tension values of:
40 Black
42 Yellow
37 Magenta
39 Photo Magenta
34 Cyan (tested yesterday)
34 Photo Cyan
39 Red
36 Green
32 Print Head cleaning solution
33 (70% Isopropyl alcohol/30% water) (pure IPA should be around 23, so this is in line)
47 Corn oil (vegatable cooking oil)

In the link given by Craig (http://www.freepatentsonline.com/5098476.html), HP claims that for their inks they want a surface tension in the range of 45 to 60 dynes/cm.

 

[Craig Ross]

Grandad
It would be interesting to see if viscosity changes on a bottle of bulk ink
over a period of time say a couple of months?

 

[Grandad35]

Craig,

The reason that I am getting into this subject is that I want to be able to do an incoming Q/C inspection of bulk inks. As was mentioned previously, there is periodic chatter about the "viscosity being off", and that this is causing printing problems. The only way to answer such questions is to actually measure the properties in question, just like is done in industry. If you were buying ink in 55 gal drums, surely you would want to test the ink in each drum before actually putting it onto paper - this was to origin of the Zahn Cup. The only unknown that we still have is what the viscosity and surface tension of our inks should be.

To this end, I took 3 empty InkGrabber carts and blew into the vent port to blow out as much of the ink remaining in the sponges as would come out without a vacuum purge - I got 7 CCs from the 3 carts. Using this mixed ink, I measured a viscosity of 2.0 centipoise and a surface tension of 36 dynes/cm - very close to my bulk ink's values. I also ran the same test on some Canon OEM Green (sucked out of a new cart), and got values of 2.2 centipoise and 38 dynes/cm.

It is interesting that all 3 inks for BCI-6 carts tested to date have similar viscosities and surface tensions. This is not to imply that these are the only two important properties for the ink, but they are two properties that we can measure ourselves.

It is my hope that others will try these tests on their own inks so that we can develop some data on a variety of inks. It would be especially helpful if you ran these tests on any "problem" inks. For example, I have seen references that imply that a low surface tension can allow the ink to be pulled back into the print head and out of the nozzles so that it takes a little while to pump the ink back into the nozzles when you start to print. This is what I believe is happening to Nifty-stuff's black - look at the vertical bars surrounding "6BK" on this image (http://www.nifty-stuff.com/gallery/inkjet-refill-1/canon_nozzle_check_pattern). I sometimes see a similar problem with Cyan on my i9900, but only on the automatic head alignment page - there isn't a problem on normal printing. I find it interesting that my Cyan has the lowest surface tension of all my ink colors (but OEM Cyan does the same thing).

Just to fill in some more data, the 70% isopropyl alcohol, printer head cleaning fluid, Windex and corn oil had viscosities of 2.3, 1.2, 0.95 and 40 centipoise, respectively. Windex has a surface tension of 39.

To answer your original question, I will re-run these tests on the same bulk inks in a few months.

 

[Smile]

I think this device would be suitable for this task:


Suitable for opaque (Pigment) Ink measurement.

Cannon-Fenske Opaque
Viscometers

Calibrated CFOC (9721-F50) Series

Model: 9721-F65
Kinematic Viscosity Range mm2/s(cSt): 20 - 100
Approx. constant, mm2/s2(cSt/s): 0.1

Minimum sample volume 12 mL (Sorry no micro version )

The price is 99$

-----------------------------------------------------------------------

To measure (Dye) ink you need:

Cannon-Ubbelohde
Semi-Micro Viscometers

Calibrated CUSMC (9721-H50) Series

Model:9722-H65
Kinematic Viscosity Range mm2/s(cSt): 20 - 100
Approx. constant, mm2/s2(cSt/s): 0.1

Minimum sample volume approximately 1.0 mL

The price is 202$

-----------------------------------------------------------------------

The only problem is the price but it would be nice if someone found a guide how to make one using insuline syringe
If you would own the real thing it is only a matter of time to calibrate custom built device.

 

[Smile]

Would this device

A standart lab pipette (dont know what size to pick) if used to count time that takes ink to rise to certain level at fixed angle would work to measure ink Viscousity?

There is also Visco Ball (Falling Ball Viscometer) technique that uses glass tube with sample and glass ball. You must measure time it takes the ball to get from one point A to point B in the tube?

 

[Grandad35]

Smile,

There are many ways to measure a liquid's viscosity. Either of the ones that you mentioned should work after you establish a baseline by testing water.

The reason that I used a syringe and needle for my tester was that anyone who has bulk inks to test obviously refills, and will therefore probably already have these items at hand. It was also fortuitous that the needles that I used just happened to have a bore that works well for ink viscosity.