What am I doing wrong? (Gordon's Reloading Tool)

After seeing the recent post about Gordon's demise, I downloaded the newest version of GRT and dusted it off -- I had it before, but I don't think I've looked at it for two years or more. I'm a pure novice at manipulating these simulators, so I ran some samples against known data (not mine) to check for correlation. I didn't find any.

First, I had lost my database of Berry's bullets in GRT, so the first thing I did was to re-enter all the 45ACP bullets, and many of the 9mm bullets. The two I generally load are the 230-gr round nose, and the 185-grain round nose - hollow base.

For known data, I looked up an article in Shooting Times that discusses those exact bullets, and conveniently used Winchester 231, which is the powder I use.
https://www.shootingtimes.com/edito...-bullet-weight-gives-edge/99399#ixzz5D6mJdC7P

I entered the data from the article for those two bullets into GRT and checked the results. They were massively different.

I know the bullet dimensions are accurate, because I entered them and I double checked them. I used the same C.O.L. as cited in the article, even though it's a bit shorter than I typically load .45 ACP. I didn't mess with the powder data, I just picked Win 231 and used the data in GRT.

GRT reported the upper level loads as being massively over-pressure. For all of the loads, in both bullet weights, GRT reported velocities ranging from 100 to 200 fps faster than what the author of the article got in real world testing.

For another data point, I looked up results from my early reloading. Using the same Berry's 230-gr RN bullet at 1.2700 C.O.L. over 5.0 grains of W231, I had a minimum of 629 fps, a maximum of 667 fps, and an average of 651 fps. GRT tells me that load should be producing 834 fps. That's a 28 percent error.

I don't expect an exact match, but differences ranging from 6 percent to 32 percent -- with most of them being around 25 percent -- were a bit more than I expected, and make the data fairly useless.

I must be doing something wrong, but ... what?
 

74A95

New member
What you're doing wrong is assuming that GRT's estimates are accurate. You've proven with your own data that GRT is wrong. It's that simple.
 

hounddawg

New member
what 74A95 said

best use I have found for these sims is to compare 2 powders for the same bullet/rifle and finding a starting load when published data cannot be found. A good example of the latter is my 30 BW for which there is no published data and my 6 ARC which has limited published data with few powders.

Like you Aguila I have found the programs to be overly conservative for pressures and overly optimistic for velocity. which is really the same thing I guess
 

tangolima

New member
I have PM'd the OP for load files and bullet dimensions.

GRT has been working for me pretty ok. Sometimes the error is a bit excessive, but I haven't seen more than 10% in speed. It usually happens with light loads, and / or immature powder models. Plated bullets seems problematic too as they tend to have more friction than cast or jacketed bullets.

-TL

Sent from my SM-N960U using Tapatalk
 

TX Nimrod

New member
Modeling software is just a tool. I use pressure modeling software in my work all the time. It is helpful in estimating what may happen, but it is not reality. Reloading software is no better - and probably worse - than loading manual data. Believe what you see happening, not what some book or computer tells you is happening.





.
 

hounddawg

New member
some of the old tried and trues come out pretty close or spot on. .260 Rem , H4350, and 142 SMK for example is within 25 FPS of a 10 round average. .308 Win with 4064 and 168 SMK data is darn near dead on.

On the other hand 6.5 Grendel, 6 ARC, .30 BW are all about 10% too high with most powders and bullets I have ran.

Irony is I wish that was reversed, it is the unknown and untested that I run the most models on.
 
If I had any assurance whatsoever that the error (or discrepancy, whatever you choose to call it) was consistently around 10 percent, and GRT was always on the high side, that would be great for my purposes. I'm never interested in pushing the limits of any load. Low to mid-range is just fine with me.

But when comparing GRT to two or three loads with known results the errors range from 6 percent to 32 percent, that pretty much tells me I can't use the software for anything other than wasting time. 10 percent is predictable (and if that varies from 8 percent to 12 percent I can live with that). A range between 6 percent and 32 percent is not predictable.
 

Shadow9mm

New member
Yeah, they will only get you in the rough vicinity. I have found some to be spot on, and some to be WAY off.
 
tangolima spotted that I hadn't changed the initial pressure from the default value, which is apparently set for a rifle cartridge. I've now used the built-in calculator to reset that, and I'll check the results against the real-world numbers.

Of course, the built-in calculator offers a choice of jacketed bullets, lead bullets, or lead (soft) bullets. I load Berry's plated bullets, so I don't know what that does to it. Berry's, of course, is no help. Their web site says to use any published data for jacketed or lead bullets.
 

tangolima

New member
I would start with jacketed. When measured data becomes available, I will use that to calibrate the initial pressure. But I would use load with higher than peak - 25% for that, as light loads are not very accurate.

There is another parameter to adjust - the bullet friction. But I would leave it alone till you become proficient with internal ballistics.

-TL

Sent from my SM-N960U using Tapatalk
 

hounddawg

New member
update

just an update on my experiments with GRT. Old saying in computer land -GIGO - garbage in garbage out.

I have spent some time to get all the variables into th eprogram and experimenting with it this last week. Simply adding in the gas port on my AR's darn near brought the model into synch FPS wise with my chrono readings. That would be in the top left hand box and was labeled Gas Leakage. I had to input the gas port size and where it was at on the barrel. That dropped the muzzle FPS dramatically. Adding the bullet jump made a small difference.

The more I am playing with the program the more I am liking it. I used the OBT (optimum barrel time) feature and got 5 shots in .5 at 100 with my AR. I accomplished that by tuning the model to real world results and inputting past chrono readings for that bullet and powder into the measurements tab.

Bottom line is in my opinion the tool works if you put a bit of time and effort into getting the parameters inputted correctly.
 
Top