If you drain & fill cold ATF out of the drain pan in a vehicle that sat over night -then a 1:1 replacement ratio of new ATF should be ok - correct ? In other words , I believe you only add LESS new ATF when the existing ATF you are replacing is warm / hot ?
Drain & Fill ATF Cold ?
- Thread starter ChrisD46
- Start date
- Status
Regardless of what the temperature of the fluid is, just add what was drained minus a quart. Then just add small increments of fluid until topped up.
I usually leave the fluid level at the "add a pint" level and drive it around until hot. Then top it off to the low side the "OK" range and call it good.
This might not work if you have one of those sealed for life AT's.
I usually leave the fluid level at the "add a pint" level and drive it around until hot. Then top it off to the low side the "OK" range and call it good.
This might not work if you have one of those sealed for life AT's.
If you're taking the view that the safe amount of trans fluid to replace is exactly the amount drained out, then yes, I think you're correct. It seems to me that both fluids would be at room temperature and thus if you replaced exactly the amount you drained out, then you should be good. The only issue I can see here is if for some reason, the fluid you are going to put in was stored at a different temperature than the actual vehicle (for example, leaving car overnight in cold garage but keeping new trans fluid in house).
I believe you're correct. Ultimately, if the fluids being exchanged are the same temperature, then you can just do the 1:1 volume replacement.
I'd personally aim about .25-.5qt low, then top off as needed when measured by the dipstick at the right temperature (for my car, they say to check when hot). This way you can be sure you're at the right level, and not overfilling or underfilling by the same amount after your D&F.
I'd personally aim about .25-.5qt low, then top off as needed when measured by the dipstick at the right temperature (for my car, they say to check when hot). This way you can be sure you're at the right level, and not overfilling or underfilling by the same amount after your D&F.
You are right. Density (therefore volume) will vary with temperature.
gathermewool
Site Donor 2025
Based on an assumed coefficient of expansion of 0.00070/*F, 4 quarts replaced, and a temp difference of 100F (60F new fluid vs a conservatively hot 160F drained fluid,) I got a difference of 0.28 quarts. This means if 4.28 hot quarts drain, then only 4 quarts of new fluid should go back in.
0.28/4 quarts is a 7% difference.
Or you could just drain the fluid cold or wait for the drained fluid to cool before measuring. Were it me, I'd add a little less than what came out hot and add the remainder after checking the volume of the drained fluid on the following day.
NOTE: I used an Internet calculator, but the numbers look good for a conservative estimate.
0.28/4 quarts is a 7% difference.
Or you could just drain the fluid cold or wait for the drained fluid to cool before measuring. Were it me, I'd add a little less than what came out hot and add the remainder after checking the volume of the drained fluid on the following day.
NOTE: I used an Internet calculator, but the numbers look good for a conservative estimate.
Last edited:
Yes go slightly under. I tried exact replacement of what I measured many times and somehow always end up a little high. Maybe they are over filling the bottles? I like to do gallon jug measurements out and replace with 4 quarts but now I'll go a little under 4 full quarts.
Drain when COLD and check when HOT
I've always done it exactly that way.
Drain it cold and replace the volume that you drained out.
It's not as though a couple of ounces either way are going to matter and it's not as though a dipstick is a high precision measuring device anyway.
That's why I always get a chuckle out of guys who make sure to top the oil off until it's exactly at the full line on the dipstick.
The full line in any two identical engines will not correspond to exactly the same volume of oil.
The dipstick is too imprecise a measuring tool.
It's merely intended to keep the tranny or engine within a safe range of fluid volumes.
Drain it cold and replace the volume that you drained out.
It's not as though a couple of ounces either way are going to matter and it's not as though a dipstick is a high precision measuring device anyway.
That's why I always get a chuckle out of guys who make sure to top the oil off until it's exactly at the full line on the dipstick.
The full line in any two identical engines will not correspond to exactly the same volume of oil.
The dipstick is too imprecise a measuring tool.
It's merely intended to keep the tranny or engine within a safe range of fluid volumes.
gathermewool
Site Donor 2025
Originally Posted By: fdcg27
I've always done it exactly that way.
Drain it cold and replace the volume that you drained out.
It's not as though a couple of ounces either way are going to matter and it's not as though a dipstick is a high precision measuring device anyway.
That's why I always get a chuckle out of guys who make sure to top the oil off until it's exactly at the full line on the dipstick.
The full line in any two identical engines will not correspond to exactly the same volume of oil.
The dipstick is too imprecise a measuring tool.
It's merely intended to keep the tranny or engine within a safe range of fluid volumes.
If checked using the same method each time, the engine oil dipstick in the my two previous Subarus, as well as my wife's previous and current Civics have always indicated full after adding the same amount of oil during an oil change.
The 97 legacy and Civic's ATF dipstick also indicated accurately.
Unfortunately, my new Subie takes an odd quantity of engine oil, so I can't dump in a 5-quart jug and call it good - i need to part of a sixth quart and check the dipstick, which for some reason is not very easy to read. The other Subies left a solid termination of oil on the dipstick, but this one seems to smear.
I guess my point is, a dipstick is pretty darned accurate, since, on level ground, it reads to the same depth into the pan each and every time - what's more accurate than that? What screws things up is us, changing how we measure, from full operating temp to slightly warmed up, and on level to sloped surfaces.
I agree with you that a few oz. won't matter much, but the difference between full on the dipstick and somewhere closer to the "add" mark can be upwards of a full quart. It usually doesn't matter even then, but more oil, at a level closer or just over full on the dipstick will add a margin of safety for unintentionally long intervals and in vehicles that consume but are not replenished.
I've always done it exactly that way.
Drain it cold and replace the volume that you drained out.
It's not as though a couple of ounces either way are going to matter and it's not as though a dipstick is a high precision measuring device anyway.
That's why I always get a chuckle out of guys who make sure to top the oil off until it's exactly at the full line on the dipstick.
The full line in any two identical engines will not correspond to exactly the same volume of oil.
The dipstick is too imprecise a measuring tool.
It's merely intended to keep the tranny or engine within a safe range of fluid volumes.
If checked using the same method each time, the engine oil dipstick in the my two previous Subarus, as well as my wife's previous and current Civics have always indicated full after adding the same amount of oil during an oil change.
The 97 legacy and Civic's ATF dipstick also indicated accurately.
Unfortunately, my new Subie takes an odd quantity of engine oil, so I can't dump in a 5-quart jug and call it good - i need to part of a sixth quart and check the dipstick, which for some reason is not very easy to read. The other Subies left a solid termination of oil on the dipstick, but this one seems to smear.
I guess my point is, a dipstick is pretty darned accurate, since, on level ground, it reads to the same depth into the pan each and every time - what's more accurate than that? What screws things up is us, changing how we measure, from full operating temp to slightly warmed up, and on level to sloped surfaces.
I agree with you that a few oz. won't matter much, but the difference between full on the dipstick and somewhere closer to the "add" mark can be upwards of a full quart. It usually doesn't matter even then, but more oil, at a level closer or just over full on the dipstick will add a margin of safety for unintentionally long intervals and in vehicles that consume but are not replenished.
Last edited:
I do this on every car that doesn't have a transmission dipstick. I never have a problem. The car I am driving right now has had this procedure done, and the transmission works perfectly.
If your transmission had some kind of seal or gasket leak, this wouldn't work, but if you had such a leak, you would already know about it.
If your transmission had some kind of seal or gasket leak, this wouldn't work, but if you had such a leak, you would already know about it.
You're positing a tautology.
Of course the same engine in the same car with the same dipstick will always give the same reading with the same volume of oil added.
Add the same volume of oil to any two of the same engines in two different cars and you'll see varying dipstick readings.
I'd expect the engine in any given car measured at the same spot on the ground to be pretty consistent.
Validity and reliability are not the same things.
Of course the same engine in the same car with the same dipstick will always give the same reading with the same volume of oil added.
Add the same volume of oil to any two of the same engines in two different cars and you'll see varying dipstick readings.
I'd expect the engine in any given car measured at the same spot on the ground to be pretty consistent.
Validity and reliability are not the same things.
gathermewool
Site Donor 2025
Originally Posted By: ChrisD46
I never knew the coefficient of expansion numbers for ATF (7%)so to round off to adding 10% less ATF then went in - once done check ATF level when hot then add as required sounds pretty full proof.
The coefficient of thermal expansion is per degree change in temperature, so that number is specific to the numbers I used. If you allow the engine a bit while you gather materiel and drain 120F ATF and add 90F ATF you've stored in your shed all August, the change will be much less pronounced.
I personally found 3 full quarts to be just a hair over the full indication on the Civic's dipstick, so I don't even bother trying to be precise. IMO, a few oz on either side doesn't matter.
If I wanted to be ultra precise and was reasonably sure that no leaks are present (or as indicated on the dipstick, and as compared to the expected amount based on manuals and Web forums,) the simplest method is to let the engine cool over night and add back in what you drain, topping off as necessary.
Calculating the expansion volume is fun and all, but not very precise due to assumptions (changing temp is huge, as well as the expansion coefficient being for oil in general not ATF specifically.)
I never knew the coefficient of expansion numbers for ATF (7%)so to round off to adding 10% less ATF then went in - once done check ATF level when hot then add as required sounds pretty full proof.
The coefficient of thermal expansion is per degree change in temperature, so that number is specific to the numbers I used. If you allow the engine a bit while you gather materiel and drain 120F ATF and add 90F ATF you've stored in your shed all August, the change will be much less pronounced.
I personally found 3 full quarts to be just a hair over the full indication on the Civic's dipstick, so I don't even bother trying to be precise. IMO, a few oz on either side doesn't matter.
If I wanted to be ultra precise and was reasonably sure that no leaks are present (or as indicated on the dipstick, and as compared to the expected amount based on manuals and Web forums,) the simplest method is to let the engine cool over night and add back in what you drain, topping off as necessary.
Calculating the expansion volume is fun and all, but not very precise due to assumptions (changing temp is huge, as well as the expansion coefficient being for oil in general not ATF specifically.)
Originally Posted By: EdwardC
I believe you're correct. Ultimately, if the fluids being exchanged are the same temperature, then you can just do the 1:1 volume replacement.
I'd personally aim about .25-.5qt low, then top off as needed when measured by the dipstick at the right temperature (for my car, they say to check when hot). This way you can be sure you're at the right level, and not overfilling or underfilling by the same amount after your D&F.
Well said
I believe you're correct. Ultimately, if the fluids being exchanged are the same temperature, then you can just do the 1:1 volume replacement.
I'd personally aim about .25-.5qt low, then top off as needed when measured by the dipstick at the right temperature (for my car, they say to check when hot). This way you can be sure you're at the right level, and not overfilling or underfilling by the same amount after your D&F.
Well said
- Status
Similar threads
