[1911_Kid]

Weighed a small piece of foam.
It started out right around 40mg.
Then i just cut the piece in half and weighed that.
All the way down to about 10mg.
On and off the scale seemed to be consistent. Was it accurate at these small weights? Dunno.
When the piece got around 4-5mg I got 0mg and sometimes 8mg, but probably expected given its a ±5mg scale.

Would this little scale last for doing 1,000's or powder measurements? Not sure, maybe, maybe not. It's not really a robust-ly made pocket scale, etc.

 

[1911_Kid]

I am sure some don't need this info, but for thise who may be looking, McMaster has decent calibration weights for cheap.

Check out these weights (use item titled "Test Weights").
https://www.mcmaster.com/calibration-weights
They come from Troemner LLC in NJ.
They come with certificate, ANSI/ASTM class 7
Each came in a bag marked with a serial #, and that # is on the cert, however, there is no Ser.No. on the weight itself.
Note the tolerances. 50g is approx ±1gr, but the smaller items have better tolerances.

I bought a 1g 5g and 20g. Scale came with a 50g. After calibrating with the 50g (which is the only way), the 20g comes in at 20.001g, 1g comes in at 1.000g, and the 5g comes in at 5.000g. Weighed each 10x.

Or for way better tolerances, spend about 3x per weight and get NIST certified, see these under title "Test Weights with Calibration Certificate".
:rock:

 

[jmorris]

Weighed a small piece of foam.
It started out right around 40mg.
Then i just cut the piece in half and weighed that.
All the way down to about 10mg.
On and off the scale seemed to be consistent. Was it accurate at these small weights? Dunno.
When the piece got around 4-5mg I got 0mg and sometimes 8mg, but probably expected given its a ±5mg scale.

Would this little scale last for doing 1,000's or powder measurements? Not sure, maybe, maybe not. It's not really a robust-ly made pocket scale, etc.

Do you have any extruded powders?

Do a test like I did in #38 and see what it does.

 

[Nitro.45]

I have a couple of RCBS for comparison look at what they can “see” vs the 505 at the end, then you can see why the charges are so accurate in #23 using the old 505, vs what you could expect from the CM 1500’s.

https://www.youtube.com/watch?v=JmxBSOOL7Ks

Cheap isn’t always bad, that old 505 and other stuff was a lot less than one of the 1500’s.

Thanks for showing that again. I think it ended the last dispute in which a member was stifled. The previous position was that a digital was far superior to the antique method of using a beam. I know we haven’t gotten to that level here, nor was it the object (no offense intended Kid) but it does put things in perspective.

 

[1911_Kid]

Thanks for showing that again. I think it ended the last dispute in which a member was stifled. The previous position was that a digital was far superior to the antique method of using a beam. I know we haven’t gotten to that level here, nor was it the object (no offense intended Kid) but it does put things in perspective.

A beam is analog, so trying to compare two measuring devices that are based in different realms of measurement is apples to oranges.

I can see where a digital may/can be much better than a analog beam. There are many factors that can make this true. Have you ever looked at how much torq is needed to make a beam move (friction in the pivot)? Some "digital" sensors can be more sensitive than the friction in a beam pivot.

I quoted word "digital" because weigh sensors are almost always an analog device, able to output voltage at some analog rate, and then the electronics inside have a bit width to read that voltage. But again, for home use, getting resolution below 1mg might be tough because the environment does not allow it to settle down. Can a beam scale measure 1mg or below?

Anyways, not here to debate what's best for powder measurement. I don't have any gun powder to weigh. I can try using talc if that is a worthy test?

 

[1911_Kid]

What am i missing form the video? The product spec says

Accurately weighs and dispenses all extruded (stick), ball (spherical) and flake smokeless powder from 2.0 to 300 grains; accurate to +/- 0.1 grain

He's trying to measure 0.2gr ???

Why would anyone expect accuracy if the device is being used outside of it's specs?

 

[USMM guy]

I have been using a Lyman 1200 DPS 3 for years now.

I am pretty happy with it. Although I am actually on my second one. Spraying one out with compressed air to clean it is a really bad idea. :dope::dope:

 

[jmorris]

A beam is analog, so trying to compare two measuring devices that are based in different realms of measurement is apples to oranges...

What am i missing form the video?

If the goal is to have charges as close to identical, to one another as possible, it’s not “apple’s to oranges”. Just two different methods to achieve the same goal. It’s worth pointing out that it can be done digitally, just costs a lot more to get that kind of resolution digitally, today. In the future it won’t though.

What you are missing is the answer to my last question in #43.

We know from the video that I posted, that the beam can detect a change of a single kernel of powder. Take your digital and see how many you can add or subtract from a given weight and it still reads the same.

The answer to that is how different your charges can be and you not know about it.

Or you can just throw volume charges from a measure and not over think it.

 

[1911_Kid]

almost all digi scales, even the cheap ones, can detect one kernel, but simply cannot display it. It would not be difficult to add a "detect light" to a digi scale where one kernel could be detected, but the weight of that is way under the accuracy spec, like my digi says ±5mg (near 0.1gr).

In that video the kernels were also dropped into the pan, which is enough force to cause the digi sensor to react and then settle down to zero.

But will ask, what was the actual weight of those few kernels? If they were less than 2gr then the results of the testing don't mean much, the scale spec says 2-300gr.

And yes, totally agree, apples-to-apples only if you are measuring withing the spec of the scale, which is 2-300gr. You trying to measure 0.2gr is apples-to-oranges for that scale.

Does the digi scale measure consistently with 2gr and 300gr from beam measures?

Let me see what my cheapo digi can do if i add 20g cal weight and then add ~4mg on top of that.

 

[BC17A]

I am sure some don't need this info, but for thise who may be looking, McMaster has decent calibration weights for cheap.

Check out these weights (use item titled "Test Weights").
https://www.mcmaster.com/calibration-weights

I bought a 1g 5g and 20g. Scale came with a 50g. After calibrating with the 50g (which is the only way), the 20g comes in at 20.001g, 1g comes in at 1.000g, and the 5g comes in at 5.000g. Weighed each 10x.

Or for way better tolerances, spend about 3x per weight and get NIST certified, see these under title "Test Weights with Calibration Certificate".
:rock:

Best way I found to sleep at night is to just make a set of known weights in the ranges we're loading to. I made a simple set of weights from some brass ranging from 1.5gn up to 5gn, in .5gn increments, and use them to verify my powder drop to scale accuracy before each loading session. Some reloaders feel the need to have the best and most expensive equipment, and that's fine, but it really isn't necessary.

 

[1911_Kid]

Here's my test.

cal scale using the 50g weight.
set scale to mg measure (display 0-50,000max)
then put on the 20g weight.
then added a tiny piece of foam to the top of the 20g.
scale consistently adds 1-3mg every time i put the foam piece on top of 20g.

but, i try to measure just the foam piece, scale stays at zero.

I not sure why this is. I do recall from instrumentation classes (way back when) that a measuring device should be used in some range between min and max of device. My digi spec says 0-50g, but for accuracy and consistency reasons I should stay between say 10mg and 40,990mg.

If i wanted to measure stuff that was in range of 5-50mg, then don't buy a scale that has a range of 0-50g, something like a 0-150mg is better.

We often overlook accuracy and consistency across the range. These tests should be done at intervals, say every 10mg. You will be surprised to see it's not a flat curve across the range.

 

[1911_Kid]

Best way I found to sleep at night is to just make a set of known weights in the ranges we're loading to. I made a simple set of weights from some brass ranging from 1.5gn up to 5gn, in .5gn increments, and use them to verify my powder drop to scale accuracy before each loading session. Some reloaders feel the need to have the best and most expensive equipment, and that's fine, but it really isn't necessary.

That is for consistency only, it does not tell you the actual weight unless your scale has been calibrated with a known good weight, and even after that, actual weight will be within the accuracy spec of the scale, like ±5mg, etc. Granted ±5mg is only about 0.1gr, which should be negligible for everyone.

So do you weigh the ring, tare it, then fill back to zero? I have yet to see what my cheapo scale does using this method.

But having a few $7 weights that come with a cert paper, should be an ok choice to make sure the scale is ok.

 

[MKeith2]

Powder scale

I have two. An RCBS digital and the Hornady Lock-N-Load Bench Digital Powder Scale 1500 Grain Capacity. I have had very good results with the Hornady. It doesn't drift and stays zeroed. Not so much with the RCBS.

 

[Twoboxer]

Here's my test.

cal scale using the 50g weight.
set scale to mg measure (display 0-50,000max)
then put on the 20g weight.
then added a tiny piece of foam to the top of the 20g.
scale consistently adds 1-3mg every time i put the foam piece on top of 20g.

but, i try to measure just the foam piece, scale stays at zero.

I not sure why this is.. . .

My own conclusion from similar tests a while back was "the scale's programming".

One of my scales, a <$20 Hornady 1500, would perform exactly the same as yours did, only ignoring the extra weight when starting at "0".

The next digital, a $125 Gempro 250, wouldn't recognize a kernel of powder (ie, a small piece of foam) either at "0" or after adding a kernel to some weight already in the pan. This scale lost zero a bit less frequently and fluttered a little less in my reloading room than the Hornady, or my Pact dispenser/scale, or my CM 1500.

My hypothesis was some programming is done in certain scales to ignore small "weight changes" under certain conditions in an effort to suppress fluttering due to noise.

 

[1911_Kid]

I not sure about programming in a differential "ignore". Perhaps maybe.

My cheapo scale paper says ±5mg. So if I start at 20,000mg and add 4mg the reading of the differential can still have error of ±5mg. In other words, the scale says "i recognize a diff here, let me measure it" but the actual "measure" part can still carry error.

As example, real actual weight of 20,000mg and 2mg
So in real true measure the weight combined is 20,002mg

But lets start with 20,000mg on the scale, then add the 2mg.
The expectation after adding the 2mg is that the scale should display the weight in the range of 19,997-20,007mg, because i need to apply ±5mg tolerance to the new reading of actual weight. But if i started with 20,000mg how could the weight go down when adding 2mg? Simple, error of the scale, and in my scale it says ±5mg.

What i do find interesting with this little scale i got is, put weight on, tare it, remove weight, the display will add the minus sign, so it can display negative. Ok, i start at zero and add a ~3mg weight, with the error applied the scale should show me something in the range of -2mg to 8mg. But it didn't. This leads me to believe that the paperwork is missing a pertinent spec, the real operating range, like what we saw on that 1500. Perhaps my little cheapo scale is not a 0-50g scale, perhaps it's more like a 0.006-49.996g scale. No way to know.

With that 1500 stating 2-300gr as the operating range, perhaps it is coded to display zero+error when the actual weight is less than 2gr? Dunno, would need to ask that manufacturer.

tare can be a kluge too because that relies on differential in the opposite direction. With mechanics, pushing down with a weight is not the same as allowing the tool to "spring" back up on it's own after applying a force.

 

[jmorris]

almost all digi scales, even the cheap ones, can detect one kernel, but simply cannot display it. It would not be difficult to add a "detect light" to a digi scale where one kernel could be detected, but the weight of that is way under the accuracy spec, like my digi says ±5mg (near 0.1gr).

The difficult part is having a load cell that can detect very small changes in mass, while being strong enough for heavier loads, then programming it so it is not always just searching, rather reads with stability. It’s easy to make the display read to .0000000 grains but useless if the numbers never stop changing.

I don’t know everything there is to know about the subject but I have dabbled a bit on a device that can sort cases to the tenth (.1) of a grain. That taught me that it’s not as easy as it looks.

https://www.youtube.com/watch?v=1V_Hm3oqlO4

In that video the kernels were also dropped into the pan, which is enough force to cause the digi sensor to react and then settle down to zero.

But will ask, what was the actual weight of those few kernels? If they were less than 2gr then the results of the testing don't mean much, the scale spec says 2-300gr

A single kernel of 3031 I was using weighs around .02 grains. The important thing to walk away with is the fact that the scales can’t resolve that amount. Add a single kernel or even a couple on any charge and it might not even change the reading. Also the fact they are constantly self zeroing to hide the fact that they drift over time. Why most laboratory equipment manuals will tell you to turn them on long before you use them to get to operating temperature before calibration and zero.

And yes, totally agree, apples-to-apples only if you are measuring withing the spec of the scale, which is 2-300gr. You trying to measure 0.2gr is apples-to-oranges for that scale.

Ok, try this with your digital scale. Put a 250gn check weight on it and see if it can detect when you add a single kernel to it.

Like this.

https://www.youtube.com/watch?v=yvF6WJs1zyY

If it can’t, how many before it recognizes that the mass has changed enough to have it read differently?

 

[1911_Kid]

Ok, try this with your digital scale. Put a 250gn check weight on it and see if it can detect when you add a single kernel to it.

Like this.

https://www.youtube.com/watch?v=yvF6WJs1zyY

If it can’t, how many before it recognizes that the mass has changed enough to have it read differently?

0.1gr = 6.48mg
0.01gr = 0.648mg
0.02gr = 1.296mg

Setting the scale to "gr" measure which only has to 0.1 resolution and trying to measure 0.02gr (one kernel) makes no sense. You at least have to get into a resolution that can display what you are after. The sensor may recognize one kernel, simply cannot display it. And if the 1500 started with 250gr weight and from there you started adding kernels, sure, expect to see no change until the # of kernels added is at least 0.1gr, because that's the smallest resolution the display has. "gr" scale to 0.1 resolution is not the right measure for weighing weights less than 0.1gr. Anything close to 1mg (kernel size) the scale has to be way better, with accuracy tolerance near +-0.5mg or better.

But I did that test using a tiny foam piece that has real weight somewhere in the 1-3mg range. Adding that to an existing 20g weight my scale did keep adding to the display, never subtracted as the tolerance allows (+- 5mg). However, when trying to just weight the foam piece by itself the scale stayed at zero.

 

[BC17A]

That is for consistency only, it does not tell you the actual weight unless your scale has been calibrated with a known good weight, and even after that, actual weight will be within the accuracy spec of the scale, like ±5mg, etc. Granted ±5mg is only about 0.1gr, which should be negligible for everyone.

So do you weigh the ring, tare it, then fill back to zero? I have yet to see what my cheapo scale does using this method.

But having a few $7 weights that come with a cert paper, should be an ok choice to make sure the scale is ok.

I first verify the accuracy of the scale within a particular range before running a batch on the press. So, say I'm reloading 380 auto with 4.2gn of powder, I'll use the 4gn ring to verify the scale actually reads 4 grains. Next I'll set an empty case along with the 4gn ring on the scale, zero(tare) the scale, then charge the case on the press and weigh it again without the ring. As long as It reads 0.1gn to 0.2 it's G2G and I'll start loading away.

 

[jmorris]

Setting the scale to "gr" measure which only has to 0.1 resolution and trying to measure 0.02gr (one kernel) makes no sense. You at least have to get into a resolution that can display

I think you are getting the point.

On your digital, that is correct. On the beam, you can detect the same mass to the kernel, or resolve the mass that accurately and the digital cannot.

That said, there are no numbers on the scale that will tell you the value despite how close they are to one another.

 

[GWB]

If you take the good ole stand by 505 and add kernels you will see the beam actually moving each time though be it very slight . I just do not understand the thought process of using a low end scale for something so important to your outcome. I understand it can be done with certain loads and not affect much when your staying in the safe zone especially pistol . But my opinion is accuracy especially in rifle depends on consistency all across the board, powder weight being a main factor . Even high end electronic scales have a slight degree of variation but to a lot less degree. In the early days I loaded 1000s of pistol plinkers with a low end electronic scale and never had a problem but over time I learned the importance of powder weight and could afford better equipment so to me its money well spent.