Myth busting - What causes digital scale zero drift and can it be fixed?

[LiveLife]

Time for another myth busting thread.

In previous digital scale myth busting threads we determined that some digital scales (even cheaper models) could be sensitive/repeatable to 0.1 gr for reloading (Actually, few were determined to be more sensitive than 0.1 gr resolution beam scales but this thread is about zero drift of digital scales so no comments about beam scales).

With more sensitive higher resolution digital scales verified by check weights down to 0.03, 0.06, 0.08, 0.1+ gr, there persists the issue of zero drift of digital scales. And that's what this thread is about.

So what causes zero drift and can it be resolved?

In the "Myth Busting Digital Scales - 1 mg Analytical Lab Scale" and "Fine tuning an 'analytical/lab' digital scale" threads and other digital scale threads, we learned that strain gauge or load cell are used for digital scales.

VEVOR Analytical Lab scale with FA DS-750, Gemini-20, WAOAW and Ohaus check weight set along with Dawn container for perspective comparison.

Load cell of VEVOR digital scale

Cause of zero drift - What I found during my resolution of Varget down to single kernel testing is that digital scale zero is influenced by internal/external factors that affect the sensitive load cell:

• Counter/bench top surface not level or stable
• Operating scale outside of manufacturer specified temperature range (typical 59F-95F but check your manual)
• Warm up of scale as electronic components/IC board temperature increase
• Scale housing construction the load cell/base is attached to
• Movement of air around the scale
• Any source of magnetic field
• Some claim fluorescent light fixture ballasts (I only use LED lighting in the house with no noticeable effect on zero)
• If battery powered, low battery
• Defective digital scale

Solution options to reduce/eliminate zero drift:

I highly recommend use of check weights:

$15 Aluminum (Down to 10 mg/.15 gr) - https://www.amazon.com/American-Wei...rds=scale+check+weights&qid=1629058966&sr=8-2

$82 Ohaus ASTM Class 6 stainless steel - https://www.zoro.com/ohaus-calibrat...edirectid=16290592283668228270010070301008005

$158 NIST Class F stainless steel - https://www.amazon.com/Rice-Lake-Stainless-Calibration-Weight/dp/B006MWG13U/ref=sr_1_2?keywords=CERTIFIED+calibration+weights&qid=1552671682&s=industrial&sr=1-2&th=1

• Stabilize and level counter/bench top surface
• Use digital scale within manufacturer specified temperature range
• Allow sufficient warm up of scale until zero drift is minimized
• Verify accuracy/repeatability of scale with check weights (NOT HEAVY CALIBRATION WEIGHTS)
• Check to see if zero drifts when four corner of more sensitive scale feet/mounting screws are pressed down - If so, adjust screw tension until zero drift is reduced or not affected.
• Close door/window/vents to reduce air movement. Use dome/cover over scale if air movement cannot be eliminated.
• Remove sources of magnetic field (In my case magnetic tip screw driver but also cell phones, speakers, etc.)
• Turn off fluorescent light fixture ballasts or use LED light source
• If battery powered, replace with fresh batteries
• If all else fails, replace digital scale (Contact customer service for replacement or get a different scale)

For my VEVOR digital scale testing, zero drift was reduced as I eliminated above listed causes with sufficient warm up of 30 minutes and removing sources of magnetic field being the biggest offenders but the zero drifted very slightly even after.

This is where I took member jmorris' practice to heart and disassembled the scale.

If you look at the above picture, you will notice that load cell is attached to the mounting plate with four screws which was crucial key to fine tuning this scale to reduce/eliminate zero drift -
https://www.thehighroad.org/index.php?threads/fine-tuning-an-analytical-lab-digital-scale.876547/

What I found was any touch or slightest pressure on the four mounting screws changed the reading along with any pressure above four corner adjustable feet of the scale. Then the light bulb went on and I realized as temperature of the scale increased in relation to the ambient temperature, there could be unequal expansion of plastic parts/base to exert uneven pressure on the load cell mounting plate.

So when I loosened the screws and tightened again until touch/pressure on the four screws/adjustable feet did not produce change in reading, my zero drift essentially disappeared.

So then I eagerly disassembled my other digital scales only to find that strain gauge/load cell were not readily accessible. So, if you have digital scale with load cell mounting plate that is accessible, you can check to see if mount screws can be adjusted to reduce/eliminate zero drift. If not, apply all the other listed solution options to reduce or eliminate zero drift.

So is this myth confirmed or busted? Do digital scales' zero drift?

Confirmed, but not necessarily due to the scale and there are many internal/external causes which can be resolved.

Only as the last resort, is the digital scale to blame and needs replacement.

 

[Blue68f100]

Good write up. The load cells mount is what I used to fix my RCBS scale.

Will add:

If your digital scales has the option of using batteries and/or external power supplies, try both. It's highly possible that each use a different circuitry. The external power supply may be run through a voltage regulator or not inside the scales. I ran into this on the GemPro. I picked up a GP500 several years ago for $15 at a gun show. I bought it knowing I had no warranty and it probably drifts. It did badly. Since I mainly use a GP250 most of the time I decided to try it's power supply on the GP500. To my surprise it no longer drifted. Problem solved. Now if your PS is 12 vdc you can use a small gel cell external battery which would eliminate all noise in the ac/dc converter.

 

[Slamfire]

You are just my type of OCD person. Great research project.

My electronic scale drifts in cold weather, and I gave up and use a balance beam when it gets cold.

 

[Blue68f100]

For those in the far north, Alaska may have electrical interference with Telluric Currents. I did a research project that required us to do field measurements in the north slop. Telleric currents can wreck havic on taking electrical measurements. So when a solar storm is going on you may not be able to get a electronic scale to stop drifting if it's not properly shielded. I had equipment that cold measure this here in Texas. So it's world wide but has no impact on equipment normally.

 

[hk940]

I don't know if scales are affected by CF light bulbs, but I had a digital caliper that when light from a CF hit it the digital display it would start counting up!

 

[entropy]

Yes, digital scales are definitely affected by EM from florescent lights. I had a two pronged approach to solving this:
1.) I eliminated the 4' florescent light over my loading bench and replaced it with a 4' LED light.
2.) I eliminated the digital scale and bought a 505.
I'll never have EM interference from the light again.

 

[FROGO207]

EM interference is a real thing and steps to eliminate it need to be used. Setting the scale on a styrofoam sheet seemed to help mine work better but not sure of the exact reason though. Also a large iron mass (press or such) near a digital scale will cause problems. This is known when using a compass already. Scale at least 3 feet away helps, six feet better if you still have problems. My bigger concern than drift is sample rate of data. Specifically when trickling a charge. Too long and you over fill. This combined with drift will cause you to loose your sanity! Or make you revert to a beam scale LOL.
Thank you on another good mythbusting post as well.

 

[Mark_Mark]

yes, but how does it compare to my Lee Scoops!

 

[Rule3]

The question on florescent light and their effect on digital (electronic) scales has been a hot topic. Most of it on reloading forums not scientific ones.
BUT the real question is, Ballasts.? Where is the research on Magnetic vs Electronic ballasts. Do the lights bulbs themselves have any effect on the scale. What distance from the scale? But there again LED is popular now, so who knows??

 

[The Glockodile]

To begin with, I seriously doubt people put their analytical balances on granite slabs (as should be done, something frequently neglected even in proper labs).

And then, there's the entire idea of micro - weighing gunpowder which barely qualifies as reagent - grade, with horrible variation from lot to lot...

Using a high - precision scale for our powder is like painting the side of a barn with a technical pen.

But we do it anyway

 

[Mark_Mark]

To begin with, I seriously doubt people put their analytical balances on granite slabs (as should be done, something frequently neglected even in proper labs).

And then, there's the entire idea of micro - weighing gunpowder which barely qualifies as reagent - grade, with horrible variation from lot to lot...

Using a high - precision scale for our powder is like painting the side of a barn with a technical pen.

But we do it anyway

I worked in a few labs. Everything is calibrated and certified by certified gauges. So unless you have a home cleanroom I don’t see why you need this level of precision. But… if it makes you happy, go for it.

 

[stormshrike]

Great info, thanks!

Remove sources of magnetic field (In my case magnetic tip screw driver but also cell phones, speakers, etc.)

Roughly how far away should magnetic sources be? Is 6-9 feet "good enough"?

 

[jmorris]

So what causes zero drift and can it be resolved?

The Wheatstone bridge is measuring resistance by balancing two legs of a bridged circuit. They can be both extremely accurate and extremely sensitive.

If you just measure values given by a load cell, they will constantly be moving, it only appears stable as we truncate the data or filter it.

Temperature change also has an effect on loadcell zero and output.

https://www.interfaceforce.com/understanding-load-cell-temperature-compensation/

The problem is resolved once someone says “that’s good enough.” but it can’t be eliminated.

 

[GeoDudeFlorida]

Q: What causes digital scale zero drift?
A: a misaligned, double-drilled Johnson rod.

 

[LiveLife]

Do the lights bulbs themselves have any effect on the scale. What distance from the scale? But there again LED is popular now, so who knows??

While every light bulb in the house has been replaced by LED bulbs, I do have a box of CF (800 lumens 9.5 watt 3000/5000K) bulbs that I can use to test whether CF/LED light bulbs have any effect on zero.

I use two clamp lights on my bench so I can use them at different distances from the scale to see if there is any measurable drift of zero. (FYI, for my "fine tuning" of digital scale where zero drift was essentially reduced to nothing, I had kitchen/dining room ceiling fixtures turned on that contained 2 each (1600 lumen 17.5 watt 3000/5000K) LED light bulbs).

how far away should magnetic sources be? Is 6-9 feet "good enough"?

I was thinking of typical bench based lighting distance: overhead, near my head and close to the scale.

yes, but how does it compare to my Lee Scoops!

I have used Lee dippers and surprisingly with certain powders, produced quite consistent powder charge weights. For me, key was the technique of back filling the dippers using reverse scooping motion.

I have tested custom dippers made from cut down brass cases with good results also.

 

[Mark_Mark]

Aliens are the main cause of drifting, especially since Planet X will be crossing Earth orbit soon. The Annuakies are back to fishish the job.

Yes, Planet X has a very dense core and will effect you scales

 

[Blue68f100]

I used strain gauge load cells and pressure transducer in the lab for years. The high end ones never drifted. What drifts the zero is actually a power supply drift. I ran 6 wire leads, 2 for voltage sense to compensate this. My equipment would not drift even over a year they were still zero and calibration never drifted. What you expect when your operating a $250k worth of equipment.

As for the new LED fixtures. They have a constant current limiting power supplies in them. I had one fail after 3 yr into a 5yr warranty. The mfg shipped out a brand new unit for replacement. Like most things electronic it seams the power supplies are the weak link.

 

[Engineer1911]

I have had experience with excessive detail or how fine is fine enough. The real question is, "Did your group sizes get smaller?"
If you can accurately and repeatable way measure nanograins of powder, are your group sizes getting smaller? If not what are you doing to improve the quality of your shooting? Measuring a "bad part" to 12 decimal places is well and good, but how do you improve the process to reduce the number of bad parts?

 

[Alex G]

Remove sources of magnetic field (In my case magnetic tip screw driver but also cell phones, speakers, etc.)

I have found this to definitely be true, especially with cell phones. For some reason my phone does not affect my scale, but when I had a friend over the other day, he walked up behind me as I was weighing a charge and the scale started to fluctuate. I asked him if he could get his phone out and showed him the variation, which promptly went away when he set it across the room. The more ya know!

 

[LiveLife]

I have had experience with excessive detail or how fine is fine enough. The real question is, "Did your group sizes get smaller?"

Very good point.

For decades, match shooters including long-range precision/1000 yard Palma shooters have won matches and set records using 0.1 gr resolution beam scales like Ohaus 10-10.

That's why I believe for most reloaders, 0.1 gr resolution is "good enough" for reloading as there are other reloading variables which could overshadow variation in powder charges - https://www.thehighroad.org/index.p...e-basics-of-the-process.892327/#post-12003540

Now, if there are handloaders and match shooters who have effectively minimized/eliminated other reloading variables to the point where more precise powder charge variance would show on target (ie member Nature Boy shooting 600/1000 yards), then more precise scales that could resolve down to single kernel of powder may be applicable - https://www.thehighroad.org/index.p...lab-scale-for-120.873830/page-2#post-11627511

So until a reloader has worked to minimize/eliminate reloading variables to show on target/reduce group size, need for higher resolution than 0.1 gr may not be warranted.

And this myth busting thread exists to support those who reach that point while confirming/busting various notions about digital scale zero drift.

 

[GeoDudeFlorida]

I have had experience with excessive detail or how fine is fine enough. The real question is, "Did your group sizes get smaller?"
If you can accurately and repeatable way measure nanograins of powder, are your group sizes getting smaller? If not what are you doing to improve the quality of your shooting? Measuring a "bad part" to 12 decimal places is well and good, but how do you improve the process to reduce the number of bad parts?

I resemble that remark!

 

[GeoDudeFlorida]

Very good point.

For decades, match shooters including long-range precision/1000 yard Palma shooters have won matches and set records using 0.1 gr resolution beam scales like Ohaus 10-10.

That's why I believe for most reloaders, 0.1 gr resolution is "good enough" for reloading as there are other reloading variables which could overshadow variation in powder charges - https://www.thehighroad.org/index.p...e-basics-of-the-process.892327/#post-12003540

Now, if there are handloaders and match shooters who have effectively minimized/eliminated other reloading variables to the point where more precise powder charge variance would show on target (ie member Nature Boy shooting 600/1000 yards), then more precise scales that could resolve down to single kernel of powder may be applicable - https://www.thehighroad.org/index.p...lab-scale-for-120.873830/page-2#post-11627511

So until a reloader has worked to minimize/eliminate reloading variables to show on target/reduce group size, need for higher resolution than 0.1 gr may not be warranted.

And this myth busting thread exists to support those who reach that point while confirming/busting various notions about digital scale zero drift.

FYI: the University has an analytic lab scale on its auction site. Bidding is up to $40.

 

[LiveLife]

FYI: the University has an analytic lab scale on its auction site. Bidding is up to $40.

Do you have the link to the auction page?

 

[FROGO207]

Back on the topic of interference the EMF might be a problem. Magnetic balasts on the other hand are completely enclosed in a steel shell so the external gauss force should be negligible. Electronic balasts on the other hand are dependent on who makes them. UL approval and good filtering or lack there of and made in China with no UL approval like a lot of light fixtures in the cheapo stores for cheapo prices. Look for a UL sticker and it will at least be up to minimal quality against fires and interference.

 

[GeoDudeFlorida]

Do you have the link to the auction page?

PM sent....