Automotive Refractometers

[Larryjb]

I thought I'd try an inexpensive refractometer. Most reviewers claimed it to be accurate, but I'm suspecting that they don't know how to check the accuracy, or I'm doing something wrong. Perhaps some one here has enough experience to set me right. If it turns out that the refractometer itself is junk, I won't be all that upset. I knew I was getting a budget tool and wanted to try it out.

I set the zero using some bottled water. I know that it is supposed to be distilled water, but I don't have access to distilled water at the moment. I could make a little to test this, but I don't think it would make much difference anyway. When I put a drop of tap water on the refractometer, the zero was very close to the zero line (32°F). It was within the thickness of the scale line anyway. I got the same result with the bottled water. No zero adjustment was necessary, although I did adjust it slightly to center the mark on the zero line.

I then got a commercial 50% mixture and put a few drops of this mixture on the refractometer. It measured as a 56% mixture, quite far off. I made my own 50% mixture using syringes and keeping everything at 70°F and got the same result, 56%.

My possible thoughts are:
1) the refractometer prism is incorrect for the scale that is printerd. (faulty refractometer)

2) The additives in addition to the ethylene glycol have the effect of increasing the actual % from 50 to 56.

3) The water used is not pure enough.

I do plan to use distilled water next, but I'm not convinced that it would cause such a discrepancy. Remember, even the commercial 50% mixture was reading the same as the mixture I made.

This difference is not so concerning when it comes to coolant. However, if the battery acid is off by the same "distance" on its scale, that makes a difference of 1.27 (full charge) to 1.24 (80%) charge.

Those of you who have refractometers, do the commercial mixture read 50%?

 

[wjburken]

Does your refractometer have a single scale for coolant or two scales, one for for propylene and ethylene?

 

[Larryjb]

I have 3 scales. One for ethylene glycol, second for propylene glycol, and third for battery acid.

I tried some propylene glycol, then discovered the stuff I had was diluted with no indication of its dilution factor.

 

[wjburken]

Wait, your refractometer reads in % of coolant, not freezing temperature?

 

[Larryjb]

Here's a picture of what I see. This is the 50% mixture.

 

[wjburken]

Here's a picture of what I see. This is the 50% mixture. View attachment 238211

Based on what a refractometer shows, which is the freezing point of the fluid, I would treat the concentration number as a reference value.

 

[Larryjb]

The problem is getting an accurate battery acid reading. The battery acid solution I tested got between 1.265 and 1.270, and the voltage reading 12 a solid 12.6 or 12.7 V. However, I'm not convinced that battery acid concentration necessarily correlates with battery voltage.

 

[wjburken]

The problem is getting an accurate battery acid reading. The battery acid solution I tested got between 1.265 and 1.270, and the voltage reading 12 a solid 12.6 or 12.7 V. However, I'm not convinced that battery acid concentration necessarily correlates with battery voltage.

A refractometer will give you the Specific gravity of the battery acid, but what a fully charged battery’s specific gravity can vary from manufacturer to manufacturer. It can also vary with the temperature of the fluid as that can affect the specific gravity.

Here is a decent explanation of specific gravity vs voltage of batteries and some of the variables.

https://batteryuniversity.com/learn/article/how_to_measure_state_of_charge

 

[Larryjb]

So, should the refractometer be reading 50% (or -30°F) for a 50% mixture? If I spent the big bucks and got a truly professional automotive refractometer, would it read 50%

 

[wjburken]

So, should the refractometer be reading 50% (or -30°F) for a 50% mixture? If I spent the big bucks and got a truly professional automotive refractometer, would it read 50%

I wouldn’t say it should. The refractometer simply measures specific gravity of a fluid and the scale converts that SG to a correlating freezing temperature. If your coolant container says a 50% mix is good to -30f the coolant may actually exceed that so they have a margin of error and don’t get yelled at by folks who don’t get a true 50/50 mix.

Might get a cheap hydrometer style tester and see what it reads. Even the cheap ones are pretty accurate.

 

[SnowDrifter]

Never found much use in hydrometers personally

And I despise the floating ball ones. They're not accurate worth a shit

Multimeter and chemical test trips is my preference

 

[Larryjb]

I agree Chris. For coolant, pH (tested using test strips) is far more important than the mixture unless you live in the far north. I got this hydrometer specifically to check battery acid as a more reliable method to assess the state of charge. I figured the easiest way to check the accuracy of the hydrometer is to check it with a 50% coolant mixture. So far, this device is precise (a huge improvement over the floating ball type), but accuracy may not be any better. On the plus side, if I can figure out how far off a 50% mixture is, I figure the battery acid scale would be just as far off and I can calculate a correction.

Of course, there are other factors that affect all this. Some batteries are designed to have a higher acid concentration. I have not seen any yet that would have a full state of charge with a SG of 1.250. I have some sulfuric acid at work and I will be able to make a solution with a known specific gravity and test it with my refractometer. I think this will be my next step.

 

[SnowDrifter]

yaadablaablaadontwannaquote

My complaint with hydrometers, and this goes for all of them; they're incredibly temperature sensitive. The consumer grade plastic stuff just isn't accurate worth a damn. I've pulled multiple samples from the same 50/50 bottle of coolant and gotten different results depending on if it's sitting in the garage or inside where it's warm. And even with that, slightly different results minutes apart.


They're.... Meh....


Frankly coolant doesn't have to be super exact. You're concerned about your pH for corrosion protection. Some other additives depending on the coolant that provide further corrosion protection (some form a layer on metal parts to protect them), and glycol concentration for balancing thermal conductivity with a resistance to both low and high temperature excursion, and lubrication of moving bits (water pump).

I run 67/33 coolant/water in my rigs for what it's worth.

Also fun little tidbit: Start your truck, dip one end of the multimeter into the coolant, another to ground.

0.25v and below - good
0.3v - look into changing it at next opportunity
0.4v - DANGER WILL ROBINSON

 

[Larryjb]

My complaint with hydrometers, and this goes for all of them; they're incredibly temperature sensitive. The consumer grade plastic stuff just isn't accurate worth a damn. I've pulled multiple samples from the same 50/50 bottle of coolant and gotten different results depending on if it's sitting in the garage or inside where it's warm. And even with that, slightly different results minutes apart.


They're.... Meh....


Frankly coolant doesn't have to be super exact. You're concerned about your pH for corrosion protection. Some other additives depending on the coolant that provide further corrosion protection (some form a layer on metal parts to protect them), and glycol concentration for balancing thermal conductivity with a resistance to both low and high temperature excursion, and lubrication of moving bits (water pump).

I run 67/33 coolant/water in my rigs for what it's worth.

Also fun little tidbit: Start your truck, dip one end of the multimeter into the coolant, another to ground.

0.25v and below - good
0.3v - look into changing it at next opportunity
0.4v - DANGER WILL ROBINSON

Interesting you say that about being temperature sensitivity. I tried a drop of hot water from the tap in the refractometer and it read 0% just like the water at 70°F. Now, what I haven't tried yet is a sample of warmed up 50% coolant. I will try this next because I am curious how much temperature will affect the readings.

 

[Larryjb]

I have now concluded that the cheap refractometers "aren't accurate worth a damn". However, I think Chris was blaming the inaccuracy on temperature control of the refractometer. I could be wrong. I personally tested mine at different temperatures, even using hot water compared to cold water from the tap. I got almost the same result at 0%. That may change at higher concentrations of coolant, but I saw now issues with temperature. Even when I used a sample of battery acid from my garage which was at 10-15°C, then warmed it up in the house to 20°C, I got the same reading. At least, that was the case for my refractometer.

I went to my lab and made up a solution of 1.341 g/mL sulphuric acid. I did this by measuring out a 10.00 mL aliquot (exact sample) of the solution, and measured it's mass to be 13.41g. This gave me 1.341 g/mL +/- 0.01 g/ml. So, the specific gravity of the solution was definitely between 1.35 and 1.3 g/mL. I forget what specific gravity value the refractometer gave me, but it was higher, somewhere around 1.37 or 1.38 g/mL. By this time I was convinced this cheap refractometer was measuring high at this end of the scale. I calibrated the refractometer so that it would read 1.34 g/mL with my sulphuric acid solution, then retested my 50% coolant. The 50% coolant was almost perfectly on the 50% line after calibration.

This tool, as a cheap tool, is useful IF you calibrate it with a 50% coolant mixture, but it will only be accurate near that range. Given that the 50% coolant line is also near the target specific gravity for most healthy batteries, it is also useful now for battery acid. It would also be useful to determine if one cell is dead, or it is also useful to test to see if the state of charge of a battery is improving while desulphating.

I would love to test truly professional version of a refractometer.

Bottom line is hundreds or thousands of customers who bought this tool are falsely believing they have an accurate tool. Even mine, after calibrating to the 50%, is not accurate through the entire scale.

 

[wjburken]

I have now concluded that the cheap refractometers "aren't accurate worth a damn". However, I think Chris was blaming the inaccuracy on temperature control of the refractometer. I could be wrong. I personally tested mine at different temperatures, even using hot water compared to cold water from the tap. I got almost the same result at 0%. That may change at higher concentrations of coolant, but I saw now issues with temperature. Even when I used a sample of battery acid from my garage which was at 10-15°C, then warmed it up in the house to 20°C, I got the same reading. At least, that was the case for my refractometer.

I went to my lab and made up a solution of 1.341 g/mL sulphuric acid. I did this by measuring out a 10.00 mL aliquot (exact sample) of the solution, and measured it's mass to be 13.41g. This gave me 1.341 g/mL +/- 0.01 g/ml. So, the specific gravity of the solution was definitely between 1.35 and 1.3 g/mL. I forget what specific gravity value the refractometer gave me, but it was higher, somewhere around 1.37 or 1.38 g/mL. By this time I was convinced this cheap refractometer was measuring high at this end of the scale. I calibrated the refractometer so that it would read 1.34 g/mL with my sulphuric acid solution, then retested my 50% coolant. The 50% coolant was almost perfectly on the 50% line after calibration.

This tool, as a cheap tool, is useful IF you calibrate it with a 50% coolant mixture, but it will only be accurate near that range. Given that the 50% coolant line is also near the target specific gravity for most healthy batteries, it is also useful now for battery acid. It would also be useful to determine if one cell is dead, or it is also useful to test to see if the state of charge of a battery is improving while desulphating.

I would love to test truly professional version of a refractometer.

Bottom line is hundreds or thousands of customers who bought this tool are falsely believing they have an accurate tool. Even mine, after calibrating to the 50%, is not accurate through the entire scale.

Thanks for sharing.

 

[SnowDrifter]

I have now concluded that the cheap refractometers "aren't accurate worth a damn". However, I think Chris was blaming the inaccuracy on temperature control of the refractometer.

I want to clarify that my complaint regarding temperature induced inaccuracies was with the hydrometers - the floating balls

 

[SnowDrifter]

I'll also add, in the professional world, we just use a multimeter and test strips as situations require. Maintenance with a refractometer was never really on our to-do list - time is just too valuable to fuss with it. Especially when you get someone with bunged out coolant and the concern is over corrosion and not necessarily the water/glycol mixture. And putting such a mixture in the thing would **** it up because it's SO bad.

95% of the time we'd just do a sniff test and check with multi meter. Coolant ratios were really only addressed by customer concern or if there was a leak and we had no idea what they've been adding.

 

[Larryjb]

As a DIY, I agree with Chris. I never used any hydrometer to test automotive coolant. In my class I set up an acid base lab to test two samples of automotive coolant: one new one and one from a car I had neglected. The kids would find the new coolant had a pH around 8 (basic, good condition), and the used coolant was around 4 (acidic, corrosive to metals). After the lab I would tell the kids where the bad sample came from, and that I had a radiator failure 6 months after I changed the coolant, and a catastrophic water pump failure at about the same time that required a tow. I also tell them they can't tell by looking at coolant. I hear so many times people say "it looks okay, no need to change it".

I got the refractometer, not to test coolant, but rather, to test my battery acid. At the very least, I wanted to see how the sulphuric acid concentration would increase as I left the battery on a pulse desulphator. When I first used the refractometer I got a value of about 1.265 to 1.270 so I thought it was in good shape. With the new calibration, that value is more likely around 1.23 g/mL. This was taken on a relatively new Costco battery that I put on a pulse type desulphator after it had been discharged a few times after the vehicle was left sitting for a month. I could not find any significant parasitic discharge, so I suspected that the battery was sulphated after sitting on the Costco shelf. After the discharge, a regular charger was put onto the battery for a day, but two cells remained in a discharged state according to the "ball type" hydrometer. I got myself this refractometer so I could get more precise results. In the end all cells had equal specific gravity (probably 1.23 g/mL).

The 50% coolant mixture can be used to calibrate the tool at the top end of the scale, then used to accurately test battery fluid. (Not many can make a standardized sulphuric acid solution very easily!)

 

[SnowDrifter]

(Not many can make a standardized sulphuric acid solution very easily!)

For what it's worth, I've found a pretty good method of desulphating a battery is to just.. Beat the shit out of it. Hook up a couple headlights, blower, whatever you want, drain the thing to 10 volts then put the biggest charger you have on it. Do that once or twice and you're in good shape. Hell, I've even noticed if I accidentally run the battery down on my car, once I have it charged up it seems to start stronger than it did before LOL. But... They key is to cram it full of charge as fast as you can. If it sits discharged you get nice little sulfur crystals which works against you.