Ken5 wrote:Even if the decent scale was a quarter inch narrower it is way too tall for the robot.

Decent website says it's 1.4cm tall. Is that inaccurate? That's actually pretty thin for a pocket scale. It's thinner than mine.

Ken5 wrote:Is a hundredths scale really faster than a tenths scale...

No. Hundredths is maybe a way to work around the thing turning itself off. Better to have a scale that just stays on. Unfortunately most of the inexpensive pocket scales don't give you that option.

The Decent scale is 101mm x 35mm. I just measured mine.

Oh, I see, we're talking about two different scales. The $99 "Decent" scale is 2mm too wide and (Ken is right) way too tall.

The $29 "Decent Simple" scale, which doesn't have bluetooth, also looks to be just a hair too wide.

Ken5 wrote: Is a hundredths scale really faster than a tenths scale, or does it only appear to be since the hundredths digit is faster?

I have bought about 4 of the cheapo ($25 or less) 0.1g scales from amazon, compared to about 4 of the cheapo 0.01g scales from amazon and all the 0.01g load cells were significantly faster at displaying weights. It had nothing to do with the tenths incrementing slower than the hundredths.

This doesn't need to be so. You can get fast 0.1g scales. It just depends on the quality of the load cell.. .. it's just if I buy the cheap 0.1g scales from amazon, I have never found their load cells to be fast responding. For a very short time, I read that Decent went through a recall of a small sample of one of its scales because its update was slow (slower load cell than they expected) ... I think all Decent scales now at 0.1g are fast to update. (I'm just using Decent here as an example of 0.1g scales, that can be slow vs fast.. all their scales now are very good from what I read)

I suspect that the problem Decent had with their "Simple" scale was about sensitivity of the load cell. There's a video posted where coffee beans are being weighed, added one at a time, and the scale isn't registering a change.

A single bean weighs between .1 and .2g, right at the level of the scale sensitivity. A .1g scale might only be able to resolve reliably to ±.1 or .2g, depending on the scale. I like having a .01g for weighing the dose (not the shot) because a .01g scale with, say, a ±.02g precision gives plenty of feedback when adding/subtracting single beans.

jpender wrote: There's a video posted where coffee beans are being weighed, added one at a time, and the scale isn't registering a change.

John, was the scale tared to zero in that video? Reason I ask is that my Maxus Brrw scale seems to have an aggressive program to keep the scale at 0.0. With the Maxus at 0.0 I can drop as many beans as I like one at a time and as long as they are smaller beans the scale never register a weight. If the scale was showing a reading other than 0.0 a single beam did change the scale. After this process of putting single beans on the scale with no change in reading the scale did show a negative value when taking the beans off, so it was taring to zero with each bean. This made no difference if the tray was empty, or had a heavy cup tared to zero, so this sort of indicates to me that it is more software than load cell issue with the Maxus.

I wonder if this 'was' the case with the decent scale.

Good news about the Maxus is that once there was weight reading on the scale even dropping a few grounds on the scale would register a reading change if it was called for.

Ken

jpender wrote:I suspect that the problem Decent had with their "Simple" scale was about sensitivity of the load cell. There's a video posted where coffee beans are being weighed, added one at a time, and the scale isn't registering a change.

A single bean weighs between .1 and .2g, right at the level of the scale sensitivity. A .1g scale might only be able to resolve reliably to ±.1 or .2g, depending on the scale. I like having a .01g for weighing the dose (not the shot) because a .01g scale with, say, a ±.02g precision gives plenty of feedback when adding/subtracting single beans.

This.

I have a $12 1000g x 0.1g scale and it's terrible for measuring out a dose. Works fine for pulling a shot, though. There's a slight delay but it's not too bad. I've seen videos of other more expensive scales and they all had either a similar delay or faux extrapolated weight that get fixed after shot stops. Acaia in particular also seems quite finicky with the way it tares/starts and I see users wrestle with it all the time on YT. I also don't see much value in its "auto start from the first drop" timer as if preinfusion time doesn't matter.

On Amazon, Weightman used to have a $12 300g x 0.01g version that's sensitive to single bean adds and also have a timer function. It worked great for me, but mine broke after it got soaked in espresso a couple times (forgot to place a cup ). Unfortunately, they stopped making those.

Ken5 wrote:John, was the scale tared to zero in that video?

No, it was at around 16g. To fix the problem the manufacturer replaced the load cell with a different one.

Timemore Black Mirror Nano. Like others, does not fit straight in, but works just fine on a diagonal.

Its espresso mode is very nice - it waits for a cup to be placed, re-zeros itself, then starts timing automatically when the first drips fall, and holds the time and weight when you stop the shot. Remove the cup and it resets to the beginning.

I have my Robot rather high up as it makes pulling the levers and looking at the naked portafilter easier, and the angled display works better for that than the top display on my other $15 scale. Honestly, the $15 ones are fine, but I found it annoying when I would forget to start the timer, and that the timer button was on the opposite side from the timer display.

I don't think this will improve my coffee, but it's definitely a huge upgrade in convenience to have a scale/timer that understands an espresso routine, and I like that the logic is self contained and not reliant on an app. I wish it were 1cm narrower of course, but it's about double to go from here to the Acaia Pyxis.