Need help with Aria accuracy

[tsnouffer]

I bought and set up my Aria about a week ago. I immeidately noticed that it was reporting my weight about 2-2.5lbs lower than my existing scale. 

Doing an experiement, I placed a 25lb. dumbell on each scale getting 10 readings each. My old scale reported 25.0 -25.1 lbs each time. The Aria reported 23.1 - 23.3 lbs. each time. Since the dumbell is a known weight, to me the Aria is about 7% below what it should be.

I've read elsewhere on these forums about how an individual's weight can vary and how the Aria can report a variance of about 2lbs. In this case, I'm looking at a static weight  in the dumbell. 

While I can understand that I guess, what I have a problem with is consistent underreporting that I am seeing.  

Has anyone else done this type of experiement or observed something similar?

Cheers

[Sabotier]

I'm having very much the same issue as many who have posted to this thread. 

I purchased my Aria scale in early May (2014), and have been using it virtually daily ever since. It stands on a tile floor in our bathroom; note that I kept our old, spring-based bathroom scale for comparison.

I can get fairly consistent readings out of the Aria, but find I need to weigh myself five to seven or even ten times in a row any given day, to get a statistically significant result (compared to a single reading on a traditional scale!). 

This morning, however, I realized that I've been led down a total primrose path. I thought I had brought my weight down to 180 lb, but on moving the Aria around on the floor just a little, I discovered I could get readings as low as 175 lb and as high as 195 lb!! (The latter actually agrees with the traditional scale.)

The Aria scale is certainly convenient, but if it isn't telling me TRUE weight figures, what's the point?

Is there some way to calibrate this thing? Any device based on a strain gauge (which I'm guessing this uses as its sensing element) needs some kind of calibration in order to tell reality.

I agree with others who have posted here - the engineering is just not mature enough for this device to be considered an accurate measuring instrument.

[MatthewFitbit]

@Sabotier What you've described is certainly outside of what we would expect for your scale's accuracy. While it may sometimes take two measurements to get an accurate reading (the first one allowing the scale to calibrate), you should not need to weigh yourself five times or more. Likewise, moving the scale on the floor should not significantly alter its measurement, unless that floor has some resilience to it. Is the floor perhaps linoleum, or some other material with give? You might try measuring on hard tile or stone, to be sure that this behavior is consistent.

If this does seem to be the case, please contact our Support Team at http://contact.fitbit.com and let them know. They will be happy to investigate and help you out - we'd very much like for you to enjoy using the scale as it was designed.

[Sabotier]

The scale stands on a hard tile floor, with absolutely no "give" - perhaps I didn't make that clear enough in my earlier post.

One thing I noticed while experimenting yesterday is that, at some locations, the scale doesn't quite have all four rubber feet firmly on the floor - I can press an edge or a corner and "rock" the scale ever so slightly. Just a tiny bit! I wouldn't notice this rocking normally; I have to get down on the floor and press the corners of the scale gently in order to determine it's happening.

I can't prove my conjecture, but this appears to make all the difference: even the tiniest amount of "twist" to the body of the scale throws the reading off (as I would expect for a strain gauge-based instrument). I've found a location where the scale doesn't seem to tip or rock, and the readings agree with those from a more traditional spring-based bathroom scale.

Mind you, I attempted a different solution that didn't work at all, because of apparent "give" - I had cut a piece of Masonite with smooth veneer on one side, and placed the scale on that before weighing myself. The value I got was about 170 lb - much lower than reality. Apparently the masonite was too soft; a piece of harder board might work better.

I would strongly suggest that the design engineers at FitBit experiment with the scale in similar fashion, to find out once and for all what are the preconditions for taking accurate weight readings. The information you currently provide is woefully inadequate.

And yes, I have contacted the Support team about the accuracy issue.

[Sabotier]

Also please notice - you didn't answer my question. Is there any way to calibrate the scale? 

[elizabethayp]

I gave up, after4 months and took mine back for a refund, to find I wasn't alone in returning. Huge shame, think you guys released it before it lives up to expectations

[Anirask]

Got my scale on Thursday and sign me up to the club of people that is getting inconsistant results with this thing.   According to the scale, my weight is going up, my lean mass is going down even though I'm under 1000 calories for my daily in/out. 

Getting a semi accurate weight readout from a scale should not be a 15 minute ordeal Fitbit.

Oh and btw - My old, non digital scale is showing me about 3 lbs lighter than the Aria.

Another day or two of this nonsense and the Aria is going back to the store.

[Warcraft]

My new Aria states that I am 60 pounds lighter than I am. Thank you!  Instant weight loss!

I tried it in several flat / solid surfaces and same deal.  60 pound difference.

Never mind trying to connect it to the network and having it actually synch.  After 30 years as a network engineer, this should have been a breeze.  Not!

I give up.  Returning it.

Same here.  After 45 minutes of getting it to connect to my network, I weigh myself and it shows a sixty pound difference.  Maybe it was all the brain matter I used trying to get this thing to work.

I am grateful I bought it through Amazon.  Going back today.

Moderator edit: merged reply

[MatthewFitbit]

@Sabotier The Aria automatically calibrates itself each time you take a weight measurement - there is no method of manually calibrating the scale. I understand that you're seeing a variation in the readout depending on exactly how and where you stand on your scale. We would expect the results to deviate only slightly as a result of standing differently, and generally recommend that users attempt to stand in a consistent position and manner in order to achieve consistent measurements. Obviously, this will involve a bit of estimation, but we do believe that the results will be sufficiently accurate to allow you to set and achieve goals.

If you feel that your scale is producing measurements that are outside of this expected range, please relay that information to our Support Team in your case. They will be able to investigate your account and offer assistance.

@Anirask Remember that weight and body fat typically don't change in proportion to one another. It is common for body fat to drop as total mass rises, or vice versa. If you are concerned about the accuracy of your device, please contact our team at http://contact.fitbit.com and let them know.

@Warcraft Sorry to hear you're having trouble - 60 pound is obviously well outside any deviation we would expect from the scale, and may indicate that there is a hardware issue with your particular scale. Make sure you have placed the scale on a hard surface (stone or tile) and not one with resilience (carpet, linoleum, wood), and that you've removed and reinserted the batteries before standing on it again. You can contact our team at http://contact.fitbit.com or return the scale directly to the retailer if this behavior continues.

[Sabotier]

Folks - 

I believe I've found the solution to this issue.

Matthew - you've totally misunderstood the point I was making. The question has to to with where the scale is placed on the FLOOR, not how I stand on the scale.

On that note - to the others who are having trouble with the accuracy and precision of this thing.

I've found that the Aria scale is EXTREMELY SENSITIVE to where it is placed on the floor. You need to make sure that all four rubber feet are on a coplanar surface. The best way to make sure of this is to get down and press the corners of the scale GENTLY to check for any rocking. You won't notice the rocking if you press hard; the plastic body will bend or twist.

If it rocks when you press gently, move it to another location so that all four rubber feet are firmly planted. Even the tiniest bit of "twist" when you stand on the scale will throw the reading off.

On a tile floor, the feet also can't "span" across the gaps between tiles.

Once I made sure of the above, I found that my scale (a) read exactly the same as my old classic bathroom scale (i.e. it's accurate), and (b) reads almost the identical value in a series of weighings (i.e. it gives precise readings).

When I submitted a support ticket with this information, the support people essentially told me what I already knew.

I'll be interested to hear if this solution works for others out there.

[christinenicole]

Sabotier (Jogger),
Thanks for your post clarifying floor positioning. I have an additional question have you found any information about storing the scale upright (leaning against a wall) versus flat? I think that is why mine reads at two different measurements consistently. Just yesterday I placed it flat on the floor to weigh myself and the first reading was 145.6, then when I immediately weighed myself again it was 141.6 (exactly 4 lbs....the variation is ALWAYS 4.2-3.9 pounds, usually 4.0).

Moderator edit: personal info removed

[Sabotier]

Christine - Yes, the bit about storing the scale on edge *IS* a point that tech support has mentioned. When you do that, the next time the scale is used, it goes through a self-calibration process. Tech support has told me that the scale can take up to two consecutive weigh-ins to recalibrate itself - that is, you would need to throw away the first couple of weight values you get after having stored it on edge against a wall. 

[christinenicole]

Awesome! Thanks! I will disregard the first reading....and find a place to store it flat!

[Anirask]

I've been using my Aria on a single flat tile as has been suggested earlier in this thread and frankly am still having the erratic results on weigh-ins.  According to the scale, I've gained 3 lbs in the last 24 hours while still very much calorie deficient during the period.

This thing is going back today.   I need a weight measurement tool.   Not a novelty that will try to guess it semi-accurately.

[jb9444jb9444]

I have been using my Aria everday for about 2 months and have sometimes weighed my self 2 or more times in the AM.  Yesterday I weighed 218 and today it was 214, now this could be possible but not very likely.  I then weighed myself again and weighed 209, 210, 209 and 214. What should I do?  I put in new batteries 3 days ago and the symbol says I have full battery power. 

[KarenMarie1234]

I've done all that and it did NOT cure the problem.  Now what?

My scales have synced to my dashboard automatically one time.  In a way, I'm greatful they haven't been able to sync automatically since then because the numbers I've gotten from the Aria are all over the place.  I've done the magic re-calibrate the scales, step on them five times, don't move them between uses, dry bare feet only.  I got frustrated this morning and weighed myself TEN times in a row and didn't get the same results twice.  In fact, my body fat consistently increased .1% every time I stepped on the scales. On top of this, the Aria measures my weight at minimum 8# MORE than any other scale.  Yes, I weigh -8# on the other scales in the bathroom, -8# on the scales at the Dr. office, and -10# on the scales at a friends house.  What a rip off.

Moderator edit: merged reply

[trentonmcanally]

I am a huge fan of using tech to assist in my weight loss and REALLY wanted the Aria to work, but in the end I had to return it.  For several days I tried to test it by stepping on and off 15-20 times and never could get a consitant weight or body fat reading.  Don't get me wrong they were all within a reasonable amount (+/-) 2 lbs, but body fat also varied (+/-) 2%.  I purchased a Taylor Scale from Kohl's and 14 out of 15 times (again several different days) it was exactly the same weight and body fat %.

[Bo3b]

Another new Aria user with accuracy concerns.

Why can't the Aria allow for a "calibration factor" via firmware? For instance, if a user gets consistently 5% low measiurements, give a Dashboard means to "adjust" the recorded weigths by +5% for that person.

[Al007]

Hello,

I was thinking of purchasing the ARIA scale. 

It appears from one of your answers, that the scale is accurate within 2 lbs.  If that is the case, I would prefer not to purchase it.  A scale's main purpose is to weigh someone (or something) accurately and consistently be accurate.  For around $120.00 or so, this scale should be accurate to within 0.2 lbs not 2.0 lbs.  You response is welcome and appreciated.  Thankyou.

[FOS]

It was so inacurate and frustrating, that I returned it and bought a Withings Smart Body Analyzer. It is spot on and never flucuates ever. I have never been more satisfied.

[randydebut]

Mine is bad too, I don't ever trust what I see from these scales. They are expensive and not knowing your real weight seems crazy to me. I love the idea behind these scales but really they are not worth the money.

The other thing is if I check my stats with my computer desktop, phone app, and android tablet app they all say different weights! 

On the verge of sending these back, I've been waiting for an update to see if this can get fixed??