.
cars

Jump to content


Hello, Allpar Forums member or visitor! If you were an active member (more than two posts) or subscriber, you would not see this ad!

Register for a free account or Sign in (see top right of page for Facebook/Open ID login icons).


Photo
- - - - -

Freezing Point Of Ethylene Glycol Antifreeze


This topic has been archived. This means that you cannot reply to this topic.
14 replies to this topic

#1 AllanC

AllanC
  • Active Member
  • 2,303 posts

Posted October 1, 2006 at 10:37 pm

Last Friday I bought several gallons of Peak Long-Life antifreeze concentrate (ethylene glycol) solution that is mixed with water to obtain proper freezing point. According to instructions on the container a 50/50 mixture (half antifreeze, half water) will give a freezing point of -34 deg F. Yet when I mix equal amounts of antifreeze and water and test with a device for measuring freezing point I only get freezing point of -20 deg F. I used a calibrated cylinder for the mixture that is commonly found in a laboratory. I know I mixed equal amounts. The antifreeze container still had its adhesive seal over the mouth when the cap was removed so it was not contaminated after it left the manufacturer. I am thinking that Peak has a quality control problem and is selling diluted antifreeze as undiluted concentrate and at a higher price. Anyone encounter such a situation?

#2 Bill R

Bill R
  • Inactive
  • 1,728 posts

Posted October 1, 2006 at 10:41 pm

Your tester must be inaccurate - find another tester and recheck.

#3 gearhead

gearhead
  • Active Member
  • 3,248 posts

Posted October 2, 2006 at 12:04 am

Tap water? Or distilled water?

#4 John Wood

John Wood
  • Active Member
  • 10,649 posts

Posted October 2, 2006 at 06:54 am

Although not a scientific way to tell, usually the concentrate will appear thicker and seem to have a slightly higher viscosity than the pre-mixed (diluted stuff).

Try warming the mixture, mixing it well, and then let it cool before testing. I would be curious to see how it tests after that.

Otherwise, as suggested, your tester may have deteriorated to the point of giving false readings. You could always buy a jug of 50/50 and check your tester. :)

#5 ckrobins

ckrobins
  • Active Member
  • 79 posts

Posted October 2, 2006 at 08:11 am

did you say you bought the antifreeze that is already mixed with water? - or the concentrate?

#6 AllanC

AllanC
  • Active Member
  • 2,303 posts

Posted October 2, 2006 at 08:15 am

Tap water? Or distilled water?


I always used distilled water in vehicle cooling systems but for this specific test I used tap water which would have some minerals in it. That would not change the density of the mixture and change its freezing point that much.

Otherwise, as suggested, your tester may have deteriorated to the point of giving false readings. You could always buy a jug of 50/50 and check your tester.


I have used a tester that had 5 small plastic balls inside. Number of balls floating indicates freezing point. That tester indicated about -10 deg F freezing point. I did purchase another type of freezing point tester. It had the name Prestone on it and uses a float with an arrow. It indicates -20 deg F freezing point. Hopefully this one is more accurate. I do know that the temperature of the antifreeze - water mixture affects the test for freezing point. Truly more accurate ones that I have seen in repair facilities have a temperature correction factor built into the gauge. A mixture at higher temperature will be less dense and will give a lower freezing point reading. I don't know to what temperature these antifreeze tester gauges are calibrated.

I will try warming and cooling the mixture some and see what the freezing point is. I did not take a temperature of the mixture when I performed the test.

Last Friday I bought several gallons of Peak Long-Life antifreeze concentrate (ethylene glycol) solution that is mixed with water to obtain proper freezing point.


The antifreeze was concentrate; not premixed/diluted.

Edited by AllanC, October 2, 2006 at 08:17 am.


#7 Volunteer

Volunteer
  • Active Member
  • 2,418 posts

Posted October 2, 2006 at 01:12 pm

So, you are basicallly saying that the 'pure' or undiluted Peak, when mixed with an equal volume of water gave a tester reading of -20 instead of the (claimed) -34?
I have mostly used the Prestone type testers and just about any brand (of glycol) shows same diluted results. You may notice that the graduated marks on the tester housing are not spaced equally. For that reason I find it much simpler to use the 'percentage' scale; ie. there are 6 marks on the (Prestone) housing which correspond to 10, 20 - - - 60% of ethylene-glycol. A 50/50 mix should be very close to the 5th. mark. My results have never strayed more than 5% either way. The temp of the liquid has little if any effect on the percentages, unlike the prehistoric alky systems.
Give this a try by using a smaller measured volume of pure stuff (ie. one quart) and then add equal amount of water. (shake well)

Edited by Volunteer, October 2, 2006 at 01:16 pm.


#8 AllanC

AllanC
  • Active Member
  • 2,303 posts

Posted October 2, 2006 at 03:14 pm

So, you are basicallly saying that the 'pure' or undiluted Peak, when mixed with an equal volume of water gave a tester reading of -20 instead of the (claimed) -34?


Yes.

The temp of the liquid has little if any effect on the percentages, unlike the prehistoric alky systems.


Temperature of the antifreeze - water mixture does significantly affect the reading that an antifreeze tester will give. More expensive and sophisticated ones use temperature correction. As liquid temperature increases it expands and occupies more space for the same mass. As the liquid temperature increases its density decreases. Antifreeze testers are measuring the density of the antifreeze-water mixture against the density of a reference standard which is the density of water.

What started this question and thread was that several days ago I drained and flushed the cooling system on my 1991 Dakota, 5.2L V8. I opened the drain at the bottom of the radiator and removed the 2 plugs on the engine block to drain as much of the tap water used for flushing. Closed radiator drain and reinstalled 2 block drain plugs. Both service manual and owners manual indicates that cooling system capacity is 14.3 quarts (for models with heater and AC) which my truck has. I added 7 quarts of fresh antifreeze concentrate. After this I added distilled water until the radiator was full. I removed the plastic overflow bottle and added equal mixture of antifreeze concentrate and distilled water to bring it to the full mark. Reinstalled overflow bottle. Started engine and let it idle until thermostat opened. After thermostat opened continued to add distilled water until radiator was full. I added about 1 gallon of distilled water though I did not measure it exactly. I assumed that there would be water in the heater core and some in other areas of the engine block which could not be drained.

Drove vehicle for 40 minutes and then let it set for 24 hours. Yesterday I performed the test on the antifreeze solution at the top of the radiator and first tester (5 floating balls) indicated -10 deg F freezing point. I was expecting something around -30 to -34 deg F. I purchased another antifreeze tester; Prestone model with float that has arrow that indicates freezing point on scale. When I used the Prestone tester, I got a -20 deg F freezing point.

Now I am suspecting the antifreeze concentrate is somewhat diluted with water. Since the seal was intact on the bottle (under the cap) the contamination / dilution had to occur at the manufacturer. I secured a 50 milliliter graduated cylinder commonly found in a laboratory. The graduated cylinder was clean and dry. I poured 25 milliliters of antifreeze concentrate into the graduated cylinder. This was the same concentrate added to Dakota cooling system. I then added 25 milliliters of tap water to the graduated cylinder. Total volume is now 50 milliliters. This makes a 50/50 mixture. I used a plastic stirring rod and stirred the mixture thoroughly. I used the Prestone antifreeze tester and got a freezing point of -20 deg F. I should have gotten something very close to -34 deg F. I did not check the temperature of the solution. I am not sure how to apply a temperature correction. But this experiment should have produced a freezing point close to -34 deg F. That is why I believe the antifreeze mixture is off specification when it came from the manufacturer.

I suppose I could complain to the manufacturer (Peak Inc) about this but I doubt if anyone would take my claim seriously. I have put antifreeze in cooling systems on all types of vehicles/tractors/etc for the last 40 years and I have never encountered anything like this.

#9 Volunteer

Volunteer
  • Active Member
  • 2,418 posts

Posted October 2, 2006 at 03:46 pm

Yes, I do see your point. If the coolant (as purchased) was indeed 'watered-down' before being sold then it would appear in more than one jug. I don't suppose you tested the first jug? - as you mentioned pouring out 7 quarts.
I have always mixed my stuff at whatever ambient temp; from freezing to about 90-ish. F. and have always attained percentages between 45 and 50. Perhaps others here can do some mixing and testing (of whatever brand) and post their results.
I would also suggest buying another (Prestone) tester, since it's only a few $ and these things don't last forever anyway.
Another method to try would be to monitor the 'pH' of the mixture using something like litmus paper (as used in swimming pool/hot-tub maintenance). Unfortunately I do not have any figures to go by except that I seem to recall the 'pure' stuff being around 10.5.
Peak is generally a good product. Can you get some other brand of concentrate and do a (50/50) mix test on that? Then go a little farther and get some pH readings (at various percentages) so you can compare with your original mixture.
As I suggested before, it's much easier to take and compare readings by using the 'graduated percentage values' as indicated on the tester housing.
Yes, we all have better things to do with our time but if I have any fresh concentrate I will also take some readings.

Edited by Volunteer, October 2, 2006 at 03:48 pm.


#10 AllanC

AllanC
  • Active Member
  • 2,303 posts

Posted October 2, 2006 at 08:55 pm

I would also suggest buying another (Prestone) tester, since it's only a few $ and these things don't last forever anyway.


The Prestone tester was purchased as a new device yesterday. The previous tester was a thin tube with 5 plastic balls. Number of floating balls indicates freezing point of the mixture. I might try to find another type of antifreeze tester and see what reading it gives on the mixture.

Another method to try would be to monitor the 'pH' of the mixture using something like litmus paper (as used in swimming pool/hot-tub maintenance). Unfortunately I do not have any figures to go by except that I seem to recall the 'pure' stuff being around 10.5.


I got some ph paper and performed the litmus test. It appears that the ph of the antifreeze concentrate is around 9 but the color variations from 8 to 10 are very close so I dont believe this method is precise or accurate enough. I did a web search and found that the ph of antifreeze concentrate averages 10.5.

I will buy another gallon of the Peak long life antifreeze concentrate but from a different retailer (not Autozone). Maybe Autozone got a shipment of Peak antifreeze that was off specification. I can drain about 2 quarts from the truck radiator and replenish with antifreeze concentrate and hopefully get the freezing point to about -30 deg F.

Thanks for your ideas. We do have better things to do with our time.

#11 gearhead

gearhead
  • Active Member
  • 3,248 posts

Posted October 3, 2006 at 09:26 am

JMO, but I doubt an inexpensive parts store hydrometer is going to be very accurate, especially they "floating balls" type.

A few excerpts about hydrometers:
Hydrometers measure specific gravity. Specific gravity is extremely temperature dependent.

Probably the most important factor in the use of a hydrometer is the temperature relationship

A high-precision hydrometer, using readings mathematically corrected for fluid temperature, can achieve a level of accuracy only within 8F of the indicated reading, even under laboratory conditions?

#12 Wizard

Wizard

    Stready as Tulip goes will win.

  • Active Member
  • 5,109 posts

Posted October 3, 2006 at 11:20 am

And know this, cannot use undiluted antifreeze, most of the heat transfer come about with water. Antifreeze prevents water from freezing too much & improves heat transfer.

The key is finding a hydrometer that is good one. I have yet to know which
I can get, comments?

Cheers, Wizard

#13 gearhead

gearhead
  • Active Member
  • 3,248 posts

Posted October 3, 2006 at 12:08 pm

Antifreeze prevents water from freezing too much & improves heat transfer.

Antifreeze actually degrades heat transfer.

"Water has twice the heat transfer capability when compared to 50% glycol antifreeze/coolant in water."- http://www.racerpart...om/redtech3.htm

Anti-freeze will raise the boiling point of water but will also effect the heat transfer capabilities of water, which will raise temperature slightly. As of now, nothing cools your engine better than plain water."- http://www.crracing....temp/index.html
(this link is dead now because C&R has a new website. They don't have their "Principles of Cooling" page loaded yet, but Im pretty sure they will have it up in the near future.)

The key is finding a hydrometer that is good one. I have yet to know which
I can get, comments?

Get a refractometer. They're much more accurate. They're also more expensive ($30-$100).

Edited by gearhead, October 3, 2006 at 12:12 pm.


#14 Volunteer

Volunteer
  • Active Member
  • 2,418 posts

Posted October 3, 2006 at 10:59 pm

I've yet to have a problem with ANY brand of anti-freeze 'concentrate', whether Prestone or Tech 2000. I'm pretty sure there aren't too many major manufacturers anyway and the (so-called) independents just purchase bulk and have it packaged to their liking.
I've also used nothing but the Prestone brand ($5.) tester and when the mix is spot on 50/50 the readout is always between 45 and 50%.
Since ethylene glycol is pretty much 'liquid plastic' it is indeed a good insulator. This is why it's never wise (or necessary) to go over the 50/50 dilution., anything stronger than 60/40 will have detrimental results.

#15 68RT

68RT
  • Active Member
  • 4,595 posts

Posted October 3, 2006 at 11:20 pm

I suspect that most of the testers used, ball and swinging needle type have quite a bit of error built into them on the safe side to cover their behinds (tester manufacturers). I have had very poor luck with the swinging needle type as they have a tendency to collect a bubble on the needle which distorts the reading and is almost impossible to get that bubble off. The ball type just says that it is below at LEAST the number of floating balls but can actually be somewhat lower. Both devices will give enough of a reading to satisfy all except the most extreme cold regions (90 PERCENT OF THE MARKET).


.
Allpar

Home · Cars · Engines · Repairs · People
Tool and Car Reviews · News

Please read the terms of use and privacy policy. We are not affiliated with Chrysler Group, makers of cars, minivans, trucks, and Mopar (TM) parts. We make no guarantees regarding validity or applicability of information, opinions, or advice. Posts may be edited and used in other parts of allpar.com and affiliated Mopar (Dodge, Chrysler, etc) car related sites. We have the right to remove or modify any message, and to ban or suspend any user without notice. Logos and site-specific information copyright © 2001-2013 Allpar LLC; Chrysler PR materials remain property of Chrysler Group.