6.5 PRC Powder Charge/Velocity Ladder Test. No Flat Spot

[Leaf Litter]

I'm mean, that's a pretty bold statement with absolutely nothing given to back it up. You should definitely call Scott Satterlee and Cal Zandt to tell them their statistics are flawed.

Don't know those people, don't really care to call them. Do whatever you want, it doesn't affect me.

And as clarification: I interpreted OP's question as looking for "velocity nodes" within a string of different charges that are all below max. The subject of a "plateau" effect when nearing max charge is a different subject in my mind and may be statistically tangible, I haven't looked into it extensively.

 

[BrentM]

Don't know those people, don't really care to call them. Do whatever you want, it doesn't affect me.

And as clarification: I interpreted OP's question as looking for "velocity nodes" within a string of different charges that are all below max. The subject of a "plateau" effect when nearing max charge is a different subject in my mind and may be statistically tangible, I haven't looked into it extensively.

That is a much more palatable response then saying anyone who looks at velocity nodes is uneducated and, well, basically stupid. Part of the appeal to this group is that there is a general feeling of respectful and educational information being shared. Internal ballistics is a highly debated subject even with science behind it. People tend to believe their method is the only method because it works for them. Have they tried other methods, maybe, maybe not. In reality many methods actually wind up with the same results. Satterlee and the velocity node, also should include seating depth nodes for throat growth, are a legitimate methods to get to an end result. I don't believe he promoted the method for any other reason than to help shooters get to an end result in a potentially faster manner. With the shortages in components many people struggled with the old methods that took a long time to master. With the labradar and magnetospeed he offered up a solution to those of us struggling to find larger lots of powder etc to burn through during the development process. For example a lb of powder has about 7000 grains of powder. If the case takes about 60 grains of powder you get 116 rounds. If you can develop a solid load in 30 rounds you still get to hunt or play with 86 rounds. This only becomes important to us when we can't get larger quantities of powder right? I think most people will make 86 rounds last a bit if needed. At least get a low round count shooter through a hunt.

 

[Beluebow]

Charge weight, seating depth, and barrel tuning by ladder test with 3-5 shot groups cannot do anything but fool you with randomness.

And all this time I've been doing it wrong.

 

[Leaf Litter]

That is a much more palatable response then saying anyone who looks at velocity nodes is uneducated and, well, basically stupid. Part of the appeal to this group is that there is a general feeling of respectful and educational information being shared. Internal ballistics is a highly debated subject even with science behind it. People tend to believe their method is the only method because it works for them. Have they tried other methods, maybe, maybe not. In reality many methods actually wind up with the same results. Satterlee and the velocity node, also should include seating depth nodes for throat growth, are a legitimate methods to get to an end result. I don't believe he promoted the method for any other reason than to help shooters get to an end result in a potentially faster manner. With the shortages in components many people struggled with the old methods that took a long time to master. With the labradar and magnetospeed he offered up a solution to those of us struggling to find larger lots of powder etc to burn through during the development process. For example a lb of powder has about 7000 grains of powder. If the case takes about 60 grains of powder you get 116 rounds. If you can develop a solid load in 30 rounds you still get to hunt or play with 86 rounds. This only becomes important to us when we can't get larger quantities of powder right? I think most people will make 86 rounds last a bit if needed. At least get a low round count shooter through a hunt.

I hear you about my "delivery," but I also stand firm on my statement. I guess my frustration lies in the fact that there is SOOOOO much bad information out there and nobody really cares to clear it up. I followed the velocity node method when I first started reloading, despite thinking that it couldn't logically make sense, and all it did was burn components and give useless data. Saving reloading components is an admirable goal, lord knows its getting more expensive. However, there's always a cheap way to do something and the most effective way to do something, rarely are they the same.

I still stand by the fact that unless you're running large number samples, your results will always be subject to statistical anomalies such as the vaunted "velocity node."

 

[Phil Rizzo]

How many times do results have to repeat, statistically, to be considered consistent? Or stable or whatever

 

[BrentM]

I hear you about my "delivery," but I also stand firm on my statement. I guess my frustration lies in the fact that there is SOOOOO much bad information out there and nobody really cares to clear it up. I followed the velocity node method when I first started reloading, despite thinking that it couldn't logically make sense, and all it did was burn components and give useless data. Saving reloading components is an admirable goal, lord knows its getting more expensive. However, there's always a cheap way to do something and the most effective way to do something, rarely are they the same.

I still stand by the fact that unless you're running large number samples, your results will always be subject to statistical anomalies such as the vaunted "velocity node."

We can agree on the samples being an issue. I see 3 shot 100 yard groups, I see 3-5 shot SD's. Not much data but it's very common. Very few people are ever going to shoot 100 rounds for data collection, especially on this page, before considering their load data to be complete. I am one of those who uses a radar to collect data and will do so on any round sent down range at a range. It's a super useful and easy to use tool for data collection. I don't mess with it in the field, unless I get the new garmin, and then I might pack it around. The last data I collected was over 200 rounds worth of information over several sessions (common practice for me). I think is very safe to say that data is enough to ascertain it's not an anomaly and founded on luck. Same goes for Satterlee and many of the top PRS shooters who follow this method. They are not messing with small sample groups. These are guys paying their bills off the shooting sports world. Not some schmuck on a forum.

 

[Leaf Litter]

We can agree on the samples being an issue. I see 3 shot 100 yard groups, I see 3-5 shot SD's. Not much data but it's very common. Very few people are ever going to shoot 100 rounds for data collection, especially on this page, before considering their load data to be complete. I am one of those who uses a radar to collect data and will do so on any round sent down range at a range. It's a super useful and easy to use tool for data collection. I don't mess with it in the field, unless I get the new garmin, and then I might pack it around. The last data I collected was over 200 rounds worth of information over several sessions (common practice for me). I think is very safe to say that data is enough to ascertain it's not an anomaly and founded on luck. Same goes for Satterlee and many of the top PRS shooters who follow this method. They are not messing with small sample groups. These are guys paying their bills off the shooting sports world. Not some schmuck on a forum.

Sample size is the only issue. Once you start dealing in large numbers those "Nodes" melt into the background noise.

 

[gregmark24]

Original posting guy here....I spent about 10 years of my career doing testing and statistics to set specifications for electrical instruments. We always strived for sample sizes of 30 or more to get the best confidence and proportion intervals. The Hornady guys in the video that I posted earlier made a very strong argument against looking for nodes. Since they have the resources to shoot a hundred or more rounds at each powder charge at a given seating depth, they concluded that time was best spent on selecting different types of bullets and powder for the best accuracy, and that there were no velocity nodes. They seated from 20 to 30 thousandths off the lands, and also out an additional 100 thousandths, which made very little difference in accuracy (don't remember about velocity). Worth noting is that if you shoot three rounds at some powder charge or seating depth that are really poor with velocity or POI goals, throw out that charge or seating depth. With my inexpensive 243 (which shoots 1 MOA at 400 yds with factory match ammo), new and once fired brass with my reloads both performed the best at 38.6 gn, both with a node and with good POI at 100 yds. At this juncture, and I no longer know why, but that's the load I'm sticking with. My buddy's 6.5 PRC had some great vertical dispersions at different charges, but with linear average velocity over powder charges and some pretty large SD's at good POI's. I think a very important final step is to take the final load (charge and seating depth) and shoot 20 to 30 rounds, perhaps 3 per target spot, and blend them together somehow mathematically (to get a larger sample size) before doing the scope zeroing.

 

[BrentM]

Original posting guy here....I spent about 10 years of my career doing testing and statistics to set specifications for electrical instruments. We always strived for sample sizes of 30 or more to get the best confidence and proportion intervals. The Hornady guys in the video that I posted earlier made a very strong argument against looking for nodes. Since they have the resources to shoot a hundred or more rounds at each powder charge at a given seating depth, they concluded that time was best spent on selecting different types of bullets and powder for the best accuracy, and that there were no velocity nodes. They seated from 20 to 30 thousandths off the lands, and also out an additional 100 thousandths, which made very little difference in accuracy (don't remember about velocity). Worth noting is that if you shoot three rounds at some powder charge or seating depth that are really poor with velocity or POI goals, throw out that charge or seating depth. With my inexpensive 243 (which shoots 1 MOA at 400 yds with factory match ammo), new and once fired brass with my reloads both performed the best at 38.6 gn, both with a node and with good POI at 100 yds. At this juncture, and I no longer know why, but that's the load I'm sticking with. My buddy's 6.5 PRC had some great vertical dispersions at different charges, but with linear average velocity over powder charges and some pretty large SD's at good POI's. I think a very important final step is to take the final load (charge and seating depth) and shoot 20 to 30 rounds, perhaps 3 per target spot, and blend them together somehow mathematically (to get a larger sample size) before doing the scope zeroing.

I've not watched the video etc but the idea to me is that they are testing something slightly different than a satterlee test. At some point I'll have to look at it. Since you started this thread I am wondering if the PRC owner would be willing to redo this test as I suggested and see if there is a different result. Reason I ask, is simply the fact he appears to be way under max and fill rate for the cartridge. There is generally a correlation between 95% and higher fill rate and best accuracy/velocity. If he approaches the max end of the charge and fill rate and finds an lower sd and good accuracy, I'd think he will find a relatively flat velocity increase in a .5 wide grain area. That is the node, most refer to, especially those using quick load for the modeling.

 

[Opa-lopa]

How many times do results have to repeat, statistically, to be considered consistent? Or stable or whatever

if you compare the "Students T" distribution (small sample size distribution) to the Gaussian distribution, when number of samples = 30, the Students T is within about 0.1% of Gaussian. What level of match/confidence is acceptable? 90%? 95? 99?

 

[gregmark24]

The 6.5 PRC friend is going to do some finer charge increments around his best vertical dispersions. Same 0.070 off the lands.

 

[Hugnot]

Pricy bullets, powder, brass, & scarce LR primers encourage smart load development.

I use the "Students T" method, being unknowledgeable about the Gaussian distribution like:



This workup gives some idea of what to expect when more ammo is made using a 95% confidence level.

I select components that provide minimum extreme spreads taking into consideration other factors such as case capacity, neck tension & bullet seating depth. A sample size of 8 or more might give some indication. Shoot another batch with same values. Don't let round cook in warm/hot chamber, after shot # 6, I paused for a cool down.

Then upon selection shoot a 5 shot group at 100, after 1 fouling round. If good, shoot some more 5 shot groups at 200 & 300.
I might get a Lab Radar. That bayonet thingy looks cool but affects groups.

Never worked with the Gaussian distribution stuff but sort of think it would be similar as "bell shaped" curve or normal distribution would also be applied. The Gaussian approach might require fewer pricy components to be used up, would tails on both ends be shorter?

 

[gregmark24]

Nice spreadsheet! Since Garmin Xero came out, there are really good deals now on LabRadar

 

[BrentM]

Nice spreadsheet! Since Garmin Xero came out, there are really good deals now on LabRadar

If you get a lab get a recoil trigger. Without one and using a suppressor life is not fun. I use the trigger 100% of the time and rarely miss a shot.