Yesterday I posted a question/poll about how much of a change in the dose of coffee do you need to make a difference in the taste of your coffee. The collective answer seems to be that you need at least 0.5g difference in the dose of the coffee to make a difference in taste, all other things being equal. This is in line with my experience, as the times I’ve made a change in dose to tweak the taste of my espresso, I’ve gone in 0.5g jumps, as changing the dose by less than that doesn’t seem to do anything in my hands.

The reason I asked about this is that my workflow in making espresso in the morning includes weighing 18.0g of beans on a super cheap scale I got from Amazon (I know, I know). After grinding, I weigh the ground coffee to make sure that there’s no retention.

What I’ve noticed over time is that the weight of the ground coffee is sometimes the same as the initial beans that I weigh. What’s more common, however, is that the weight of the ground coffee coming out is 0.1-0.3 g more than the weight of the beans I put into the grinder. The ground coffee never weighs less. I kept track of this, and over the course of last week I’m up by 1.5g.

I have only 2 explanations for this.

  1. Although my scale displays the weight to 0.1g of accuracy, repeatability to 0.1g doesn’t happen. (You get what you pay for.) When I’m grinding my 18.0g of beans, 18.0g of ground coffee is coming out, but the scale gives me 18.2 because of the repeatability issue. This seems to be likely, as when I weigh the dosing cup, take it off the scale, and put it back on, the weight bounces between 109.1-109.3g. I also think that from one morning to the next, the scale is accurate enough, as I consistently get the espresso that I want with the 18.0g of beans I’m weighing out to start.
  2. My grinder is magically creating coffee each time I use it. This would be a violation of thermodynamics, however. The plus side of this scenario is that I’ve solved global warming. 😆

The reason I asked about how much of a change in dose makes a difference in taste is that I wanted to gauge how much of an issue this really is. Whatever my scale is doing, it seems like it’s within 0.1-0.2g in terms of repeatability. This means that this inaccuracy may be a moot point if it takes a larger change in weight to change the taste of the espresso.

The main thing I use my scale for is to see if I have retention during the grinding process. If the weight of the ground coffee is consistently higher than what I first weighed out with the beans, and I use the bellows on my grinder until I don’t see any more coffee coming out, I’m pretty sure I’m consistent with getting to the zero retention point. The few times I’ve taken apart my grinder for cleaning bears that out, as I’ve never seen big collections of coffee inside the grinder.

My OCD side is telling me that I need to get a new, more accurate and repeatable scale for making my espresso because it has this error of 0.1-0.2g.

My practical side is telling me that because (1) I can consistently make excellent espressos from my wife and myself every day, and (2) it seems that the collective experience is that it takes at least a 0.5g change in the dose of coffee to change the taste of espresso, that the scale is good enough for this purpose.

Am I missing something here? Should I get a new scale anyway?

    Which grinder have you? Also, are you catering for scale drift? If you leave it on for a while tared, sometimes it will drift up and down a bit.

    So, before you do and replace your scale (mine is about 10 years old, and costs approx. £4.50).

    Turn your scale off;
    Tare the container;
    Put the coffee in ( say 18g)
    Note the weight of the container;
    Grind into the container;
    Weigh the container with the ground coffee in;
    Subtract the weight from the container.

    Did it increase?

      wilburpan Although my scale displays the weight to 0.1g of accuracy, repeatability to 0.1g doesn’t happen. (You get what you pay for.) When I’m grinding my 18.0g of beans, 18.0g of ground coffee is coming out, but the scale gives me 18.2 because of the repeatability issue.

      Repeatability is as important as accuracy. You say your scales are accurate to 0.1g, do you mean this is their stated accuracy, or do you mean that they read to 0.1g resolution (many scales that read to 0.1g are only accurate to 0.3 or 0.4g).

      Weigh the same dose a few times on the scales, what variation or you getting?

      Most “zero retention” grinders are only able to dose to 0.3g to 0.5g (measured with 0.01g scales) over a good sample of grinds. The idea being that most folk would only weigh in and not think about it if their dose consistency is 0.5g or less.

      If your espressos are consistently excellent and you need to make deliberate changes to change the flavour, I don’t think you have a problem.

      I dose 16.5g-17g, grind, transfer to the PF, distribute and tamp. I do not weigh the ground coffee. I think you are over thinking. As others advised you, if the coffee is excellent, just enjoy :-)

      Same, use my yuaga scales that are nearly 3 yrs old, 17g grind and tamp. I don’t weigh ground either

      Decent De1pro v1.45 - Niche Duo - Niche Zero - Decent is the best machine ever made -

      wilburpan Should I get a new scale anyway?

      I like things that look purty, so that would be hard hell yeah, get a new one 😀

      Seriously though, if it’s accurate to within 0.1g it’s fine. But once you start getting up to 0.3g variance, that would be my cut off tolerance point.

      I have a Timemore Nano. I couldn’t bring myself to spend the $$$$ on an Acaia Lunar.

      Prior to that, I had an inexpensive scale. However the scale was slow to react to / display charges in weight, making it difficult to stop a shot at a precise output weight.

      The Timore is not well designed, but it is accurate and fast. I use it in it’s basic tare and weigh mode, as it sometimes stops weighing prematurely in espresso mode.

        MediumRoastSteam Which grinder have you? Also, are you catering for scale drift? If you leave it on for a while tared, sometimes it will drift up and down a bit.

        So, before you do and replace your scale (mine is about 10 years old, and costs approx. £4.50).

        Turn your scale off;
        Tare the container;
        Put the coffee in ( say 18g)
        Note the weight of the container;
        Grind into the container;
        Weigh the container with the ground coffee in;
        Subtract the weight from the container.

        Did it increase?

        I have a DF83.

        My usual process is pretty much as you describe. I turn on the scale, put the cup that collects the ground coffee onto the scale, and tare it. The scale reads 0g.

        Then I weigh out 18.0g of coffee beans. The scale now says 18.0g with the the beans in the cup.

        Then I put the beans into the grinder, put the cup in position to collect the ground coffee, and grind away. I use the bellows until I don’t see any coffee coming out from the grinder. Also, the grinding noise has stopped by that point. So I’m pretty sure all the coffee that’s going to come out has come out.

        When I put the cup with the ground coffee in it back on, the weight will often be 18.1-18.2g, sometimes a little higher. But never lower.

        I’m sure that the increase is not due to retention, as I should have lost some coffee when grinding at some point. But that never happens.

          Thanks for the input, everyone! I do think my scale reads out to 0.1g, but repeatability is to within 0.2g. And given that the dose of coffee needed to change the taste seems to be at least 0.5g, I’ll just enjoy the espresso that I’m making and not worry about it.

          Heck, maybe I’ll get enough confidence so that I don’t weigh the ground coffee anymore. 😆

            wilburpan Heck, maybe I’ll get enough confidence so that I don’t weigh the ground coffee anymore. 😆

            It’s a tough habit to shed. My beans are pre-weighed into single dose sealed glass jars. Yet I still weigh my beans (in the Niche cup) before grinding. And then I tare my basket/dosing collar, grind and weigh the grounds. If I haven’t let my NZ run long enough (to get that last 0.5g out), I’ll find I’m short. So I do try to let it run. Then I weigh yet again and sometimes (or more often than sometimes) need to use a cute little bellows to eek out that last 0.1 or 0.2 g. Verified by yet another weigh. It’s more than a bit OCD. 🤪

            But….if I simply dumped my pre-weighed dose in, let the NZ run sufficiently long, then pushed the bellows once, the end result (I know, I know) will be the now ground complete dose. Xg in, same Xg out. Good enough and I will never notice the taste difference anyway.

            wilburpan I’m sure that the increase is not due to retention.

            I understand what you are doing. But you didn’t follow my suggestion of not zeroing out the scale first when measuring the weight of ground coffee.

            Make a note of the weight of the container where your ground coffee ends up. Grind into it as you usually do. Put that on the scale. Subtract the weight from the container. Does that still read over the 18g you put in?

            I’m trying to assert your scale is not the problem by not zeroing it out first.

              MediumRoastSteam Knowing something about how scales work, I don’t see how zeroing/not zeroing would affect the scale’s behavior. The zero function just adjusts the zero point, which is an electronic function. Inaccuracies in repeatability are far more likely ro be related to whatever mass-sensing mechanism is used in the scale.

              Be that as it may:

              Turn your scale off; – Done
              Tare the container; – not sure what you meant by that, but this morning my container is 109.1g. I’m not hitting the tare button like I usually do.
              Put the coffee in ( say 18g)
              Note the weight of the container; – 109.1g + 18g = 127.1g
              Grind into the container; – you didn’t ask for this, but I put the beans into the grinder, and reweighed the container before grinding. Now it’s 109.0g.
              Weigh the container with the ground coffee in; – 127.3g
              Subtract the weight from the container. – again, not sure what you mean here, but I transferred all the coffee into the basket. When I put the container back on the scale, it’s 109.1g

                JHCCoffee Prior to that, I had an inexpensive scale. However the scale was slow to react to / display charges in weight, making it difficult to stop a shot at a precise output weight.

                I have a 9Barista, which means that the extraction I get is the extraction I get. So I don’t need real time readouts on weight for timing the extraction.

                Which makes my repeatability issue with my scale a bit more frustrating, as I just want to know the weight of the ground coffee coming out of my grinder. But since it seems like that doesn’t matter, that will save me from buying a new scale.

                I do think that if I had a conventional espresso machine, this scale would have been in the garbage by now.

                wilburpan - So it added 0.2g again… Interesting… But let’s not give up:

                Try a different experiment…

                Instead of putting the whole beans into the grinder and grinding them, but them back into the destination container over 15 seconds or so. In a way, pretending you have ground the coffee. :-)

                If you don’t get the exact same weight as your input weight, (i.e. it’s not adding the extra 0.2g) I’d be inclined to say there is no other explanation apart from grinder retention.

                If you do get a weight variation, then your scale is definitely bonkers.

                  MediumRoastSteam Dude. It’s the scale. If there was retention, and the scale was accurate, at some point I would’ve lost some coffee due to retention, and the scale would have told me so.

                  But the fact is that the scale never reads below the initial weight of the beans that I put into the grinder. As I said previously, I kept track of the weight after grinding over the course of one week, and by the end of the week I was up by over 1g. That’s not retention.

                    MediumRoastSteam Probably not, given that the repeatability error is smaller than the amount of change in the dose that seems to be required to cause a noticeable change in the taste of the espresso.

                    Just for kicks, I ran another set of weights according to your suggestion.

                    1. Weigh the container – 109.0g. Note that this is different than the 109.1g that the scale usually reads. I didn’t mention this before, but the scale says that the container weighs between 109.0-109.2, depending on the day. Most of the time it’s 109.1g, but there is some variation here.

                    2. Add beans until the scale reads 109.0g + 18g = 127.0g.

                    3. Take the beans out. Now the scale reads 109.1g.

                    4. Wait a while, and then add the beans one by one back into the container. When I finished, the scale reads 127.1g.

                    You guys are working in 0.0 of grams

                    Personally I really don’t think it matters that much

                    If you scales are wandering like a clock fair do but in coffee I don’t think it matters.

                    At work we mix paint in the 0.0g range and we even have a % of error were we don’t have to re calculate the mix.

                    Decent De1pro v1.45 - Niche Duo - Niche Zero - Decent is the best machine ever made -

                    wilburpan While it is up to you if you would like to splash your hard-earned cash on a new scale or otherwise, would you be able to tell a difference on blind tasting if you were to make two cups with 0.5g/1.0 difference between them? I assume all other parameters are the same.

                    If you can’t, why bother? If you can, you are a super taster and you are going to lose some decent cash. I assume, you can get away spending money. 😂

                    Yuaga scale all you need £17 😂 mine are 2 yrs old

                    Decent De1pro v1.45 - Niche Duo - Niche Zero - Decent is the best machine ever made -