DF83v vs DF64v (I have both) - Page 3

Grinders are one of the keys to exceptional espresso. Discuss them here.
Kran
Posts: 236
Joined: 6 years ago

#21: Post by Kran »

Yes, RDT.

Also, I don't use the bellow. With the two piece magnetic chute just lift the top portion and release. It clacks and gets whatever is stuck in the chute out. Should also mention I cut out the declumper flap. Straight though to the burrs for me.

HobWash (original poster)
Posts: 18
Joined: 2 months ago

#22: Post by HobWash (original poster) »

Jonk wrote:What's the glass for?
That's where I keep my coffee cup hahaha

HobWash (original poster)
Posts: 18
Joined: 2 months ago

#23: Post by HobWash (original poster) »

UPDATE 02/26/2024 -- It's going back

Unfortunately, I am convinced I have a faulty grinder. My hope is that someone else will do similar tests and report that their DF83v is fine and prove that mine is just a one-off defective one.

Everything I said in my original post regarding the design is still true. However, the inconsistency I experienced with the SSP Lab Sweet Burrs is also happening with the DLC Espresso Burrs.

Keep reading for the geeky details...

Test Approach

- Gear -
I have a Breville Dual Boiler. It is PID Temp Controlled. I measured the head with a thermal camera and found the temps to be super consistent regardless of the number of shots. I am set to 197F and observed 196-199F repeatably.
I also have a DIFluid Refractometer which I find to be repeatable with some care.
DF83v set to 1000 RPM
DF64v set to 1000 RPM

- Test Prep -
I individually dialed in both grinders to hit first drips at 11 seconds, then peak at 9 Bar and produce the desired yield at 32 seconds with the same beans. I then ran a 1/4 pound of those beans through each grinder straight into the trash to rule out any grinder "settling".

- The Test -
For a total of 20 shots, Shot-1 DF83v, Shot-2 DF64v, Shot-3 DF83v, Shot-4 DF64v .. repeat until Shot-20.
Observe the time it takes to produce the desired yield. We're interested in the shot time standard deviation for each grinder.

Why? I did it this way to rule out changes in the espresso machine temperature, or humidity, ambient temps, etc.
The theory is, if something changed in the espresso machine or environment, then we'd see it reflect in shots from both grinders.
Whereas if I had pulled 1-10 from the DF83v and then 11-20 from DF64v, one could argue that the espresso machine had "warmed up" during the 1-10 and benefitted the 11-20. So this method rules that out.

- The Beans -
I bought 10lbs of beans that appear to be very consistent from bag to bag, they all have the same roast date and are relatively fresh, so this is about as ideal as I can get for testing. I will not mention which beans because I do not want to stir up any debate. Suffice to say, my test eliminates any variation in the beans (keep reading).

Test Result
Some quick commentary:

Normally I will use my slayer-mod to keep the Bar @ 9 then taper down to 7 around 20s, but for the sake of science I did not touch it.
First drips did not vary much between either grinder, +/- 1s, except for Shot-7 which I stopped because it took 18 seconds and spiked to 11 Bar and it sounded like it was hard on my machine.

As for the TDS observed with the DIFluid Refractometer. I did not test every shot, that would take forever.
Instead I tested a 1 norm and 1 high from the DF83v and 2 norm from the DF64v, indicated below.
I typically observe TDS +/- 0.2 on my refractometer and target about TDS of 9.5.

I am rounding the time to the nearest second.
I am rounding the pressure to the nearest 0.25 Bar as observed around the 5-10 second peak before it starts to naturally drop.

Shot-01 DF83v - 32s, 9.50 Bar
Shot-03 DF83v - 27s, 8.50 Bar
Shot-05 DF83v - 31s, 9.25 Bar
Shot-07 DF83v - Fail 11.00 Bar
Shot-09 DF83v - 34s, 9.50 Bar
Shot-11 DF83v - 32s, 9.50 Bar TDS = 9.3
Shot-13 DF83v - 35s, 9.75 Bar
Shot-15 DF83v - 44s, 10.50 Bar TDS = 9.8
Shot-17 DF83v - 29s, 8.75 Bar
Shot-19 DF83v - 38s, 10.00 Bar

Shot-02 DF64v - 32s, 9.00 Bar
Shot-04 DF64v - 33s, 9.00 Bar
Shot-06 DF64v - 31s, 8.75 Bar
Shot-08 DF64v - 32s, 9.00 Bar
Shot-10 DF64v - 33s, 9.00 Bar
Shot-12 DF64v - 33s, 9.25 Bar TDS = 9.4
Shot-14 DF64v - 32s, 9.00 Bar
Shot-16 DF64v - 31s, 9.00 Bar TDS = 9.5
Shot-18 DF64v - 32s, 9.00 Bar
Shot-20 DF64v - 33s, 9.00 Bar

Thoughts...

My ability to stop the shot at the desired yield is (at best) +/- 0.5s. Which gives us a 1 second swing attributable to human error and scale lag. However, we are using the same human and same scale for both, so if both grinders are equal, we should see similar results. We do not.

Note that I rounded to the nearest 0.25 Bar. Though we see a lot of 9.00s, the DF64v does vary by +/- 0.1 Bar.

Obviously there is a correlation between Bar & Time.

It appears the DF83v has a wider deviation and some occasional mistakes.
It appears the DF63v is shockingly consistent.

DF83v seems to be +/- 3.0s with a few outliers. The outliers tend to be high, not low. I suspect the grinder sometimes produces extra fines that clog things up a bit more than usual? That is a total guess. What do you think?
DF64v seems to be +/- 1.0s with no outliers. NOTE: Even though we give a 1s swing for human & scale error, I did observe a slightly larger delta in Bar that coincided with the time. So I would say as at least some of that 1 second swing is due to the grinder, but still amazing.

Having said all that, I do not taste a wildly different flavor from shot to shot with the DF83v, it's not clear to me that the +/- 3 seconds makes any notable difference. What bugs me is the few outliers.

malling
Posts: 2923
Joined: 13 years ago

#24: Post by malling »

How many kg have you had through the DF83v burrs?


Yes df copy variances is like winning the lottery you might get one perfect but then again you might not. It's not exactly first time someone experiences massive play, changing alignment through the grind setting and really bad design decisions.

tompoland
Posts: 269
Joined: 3 years ago

#25: Post by tompoland »

I have to say a big THANK YOU for the time and effort you've put into this.

I have the DF64V and a DF83V is on the way so I can conduct a review, and your post is an incredibly helpful heads up on the issues I need to look at and compare.

Thanks again.
A little obsessed.

tompoland
Posts: 269
Joined: 3 years ago

#26: Post by tompoland »

HobWash wrote:UPDATE 02/26/2024 -- It's going back

Unfortunately, I am convinced I have a faulty grinder. My hope is that someone else will do similar tests and report that their DF83v is fine and prove that mine is just a one-off defective one.
Sorry to read that. I'll try and remember to let you know how I go when mine lands in about a month. Following your updates with interest.
A little obsessed.

HobWash (original poster)
Posts: 18
Joined: 2 months ago

#27: Post by HobWash (original poster) replying to tompoland »

Glad to hear you are sticking with it. I would hate to cause anyone to have a knee-jerk reaction and assume "The DF83v" is bad. I very strongly suspect this is a one-off defect. DF series have proven to be solid grinders and I expect if there is an "issue" that the manufacturer will send out a fix as they have done in the past. I'm a big fan.

In my case, I just happen to have a 2nd DF64v here already, and I like it -- so I've decided to just go the convenient route of wielding double DF64Vs ;)

HobWash (original poster)
Posts: 18
Joined: 2 months ago

#28: Post by HobWash (original poster) »

malling wrote:How many kg have you had through the DF83v burrs?


Yes df copy variances is like winning the lottery you might get one perfect but then again you might not. It's not exactly first time someone experiences massive play, changing alignment through the grind setting and really bad design decisions.
As of now, about 5kg through the DF83v. It only took about 1kg for my DF64v to settle in.

Yeah, I think there is some higher probability for defect earlier in the manufacturing for a new model, especially with how quickly they roll out new models.

My DF83v is very very low serial number.

Both my DF64Vs are newer generation (with the updated controller) and have been flawless.

So my hope is that, I'm just experiencing an "early adopter" issue ;)

tompoland
Posts: 269
Joined: 3 years ago

#29: Post by tompoland »

Yes sticking with it. It's what I do most months, try a new grinder and then if it's not a keeper, on sell it here in Australia on the coffee snobs forum.

I'm with you on the DF64V, slim, small footprint, variable RPM. It's hard not to like. And I also have two of them.

I'm also quite impressed with the Niche Duo which is the standard by which I'll be measuring the DF83V.

Thanks again for this thread, I know that there is a lot of work in keeping records like you have.
A little obsessed.

HobWash (original poster)
Posts: 18
Joined: 2 months ago

#30: Post by HobWash (original poster) »

One more test...
It turns out I over estimated the amount of coffee needed to make a decision. So I decided to use the remaining coffee to do some more tests.

I have two DF64Vs and my earlier test only included one of them for comparison with the DF83V.
However, wouldn't it be fun to see TWO data samples of the DF64V to see if the earlier one was a fluke?

It wasn't a fluke. My 2nd DF64v is even more consistent... Honestly, these results are just getting silly.

Same RPM, 1000.

Results from my 2nd DF64v
Shot-01 DF64v - 32s, 9.00 Bar
Shot-02 DF64v - 30s, 8.75 Bar
Shot-03 DF64v - 30s, 8.75 Bar
Shot-04 DF64v - 30s, 8.75 Bar
Shot-05 DF64v - 31s, 9.00 Bar
Shot-06 DF64v - 30s, 8.75 Bar
Shot-07 DF64v - 30s, 8.75 Bar
Shot-08 DF64v - 30s, 8.75 Bar
Shot-09 DF64v - 30s, 8.75 Bar
Shot-10 DF64v - 30s, 8.75 Bar

Same rounding as the earlier test. Which is no longer high enough precision to adequately capture the incredible consistency of this grinder.

So to better articulate, I will just say they all hovered around "just under 9" except for the first one which was "just over 9" and Shot-05 which was "right-on 9". I suspect the first shot may have had some left over grinds from an earlier setting resulting in some extra "fines" but I cannot be sure.

If we choose to omit the first. Well you can see, the shot times are unbelievable.

Two interesting thing to note.

1. This DF64v has the stock DLC Burrs because I do pour over on occasion. I cannot say if this had any impact on the consistency. If I were cooler, I'd swap the burrs between my two grinders and see if it affected the times, but doing that would not be fair as these DLC burrs were so perfectly aligned for this grinder straight from the factory that putting them in another grinder may not have the same magic and yield misleading results.

2. This DF64v came with the "Slow Feeder" upgrade where my other DF64v did not. I cannot say if this had an impact on the improved consistency but... There you go. I will say that in all the tests, I was just dumping in the beans for the sake of consistency. So this is not an advertisement for the slow feeder upgrade, but rather, a suggestion that feeding your beans slower **might** actually make a notable difference.