Compak K10 WBC has landed! - Page 10

Grinders are one of the keys to exceptional espresso. Discuss them here.
walt_in_hawaii

Postby walt_in_hawaii » Aug 04, 2018, 12:42 am

Sigh... lots to figure out.
I'm posting because its been about or 4 or 5 minutes on this pull and the lever is still at 9 o'clock :(
...so yeah, it can choke the machine now. The 15 thousandths was more than enough off the top of the new chamber walls. However, I also just got a false 'bottom' reading... I had the turret a couple turns too coarse and ran beans through it.. immediately spotted the coarseness of the grinds and realized it was like pourover sized.... adjusted the turret finer and it went down, grounded (I thought) and I backed off.... huh? looked too coarse; sure enough, I got a gusher on the next pull. But I thought it was just off from bottoming? took it apart... nope. The coarse grounds were on top the grind chamber shelf and I had touched down onto the grounds, not the shelf, squeezing the coarse grounds. cleaned it up and touched down/backed off and sure enough, choked the machine... ok, lots to figure out. The stock setup does not have a 'false bottom' because there is no bottom; but this doesn't really bother me, as in reality it will only ever stay waaaay over on the fine espresso side, so all I will need is adjusting from 8g to ~20g loads, which is not a large traverse of the turret and will always be on the very fine side, so I can't get those coarse grounds plugging up the top shelf again. Whew.

walt_in_hawaii

Postby walt_in_hawaii » Aug 04, 2018, 4:09 pm

I got a few pours out of the k10... boy do I love the taste of this grinder! Definitely different from my mazzer SJ (with SSP burrs). The SJ has more top citrusy flavors, but the same beans in the k10 have a more mature, evenly rounded flavor, more midrange and bottom richness, much more satisfying in my opinion. Then.... it froze again :(
Disassembly... it looks like what I had feared in the beginning is happening. The delrin is just too soft, it flexes a bit too much. Normally that's not a problem, because it springs back. But in this instance, with very lightly roasted beans, the grinder is under a bit of pressure and the inner burr is moving slightly... the center tightening/holding screw for this burr is reverse threaded, against the direction of rotation; so when the burr shifts, it serves to actually tighten the screw more, a self-correcting feature. As a general rule I don't like gorilla-ing my bolts and when I make errors, its usually leaving the screws a tad too loose. This has come from a lifetime of OVER tightening things and stripping them, and then having to fix THAT. Much easier to just tighten a bit more, ya know.
Anyway, in the stock design there is the aluminum body of the spider to tighten down against. In my modded design there is the delrin spider, which is low and flat, and the inner burr against it presents a wide, flat surface with very light loading, so it won't squash the spider at all.... but the delrin bushing, on the other had, is long and thin, and all that pressure is supported on the bottom of that bushing; so I examined that part closely, here is a pic:
Image
can you see the ridges pressed into the bushing? this is the very bottom of that bushing, its upside down in the pic. This part rests against the stock spindle's shoulder and must support all the weight of that screw, and the turning torque exerted by the burrs... obviously its failing to do so and is turning and compressing. Too soft. I don't have brass in this size, ordered some off *bay this morning but don't want to wait for it to come in... this is if I can't figure it out and have to make the bushing in brass. But for now, I've got some ideas that might solve this... its saturday morning, time to get busy in the garage.

ira

Postby ira » Aug 04, 2018, 5:01 pm

Seems like you should just just put a metal sleeve inside the delrin. Then you get the best of both worlds

Ira

walt_in_hawaii

Postby walt_in_hawaii » Aug 04, 2018, 8:31 pm

Ira, right on :) you read my mind! guess what I've been doing all morning?
Oh, and it needed a design change... I put the aluminum flange on the TOP of the bushing instead of squeezing the delrin spider in between the bushing and the burr, so the burr will bear down on the aluminum bushing directly, just as it does in the stock design (burr to spider in that design, but in mine, I have the grind chamber insert in the way, so have to have the bushing through it)... which is probably why the first aluminum bushing galled, come to think of it. Ran fine when no beans were in the chamber... but soon as the beans were added, they present enough resistance to move the burr back and crush the delrin. Now, it will crush down on the aluminum, which doesn't give nearly as much as delrin does.
Image
Image
Image
Image
Image
Image

walt_in_hawaii

Postby walt_in_hawaii » Aug 04, 2018, 8:46 pm

But... (didn't you know there was going to be a but?)
I'm a cheap bugger, and its starting to show. The amount of space in the middle hole was limiting the size of the aluminum bushing pressed into the delrin outer bushing, so I widened it. But, now the 2 bushings pressed together are very nearly the same size as the flange on top. I limited the flange to just under 1" diameter, because I have lots of aluminum in 1" diameter, its a popular size to make things from and is massively cheaper than the next size I have, 1 1/2". But now the flange is much too small, and the weakness of my design is starting to show. What is happening is, the rotation torque/sweeping power of the spider is ONE component and the SECOND component is the vertical crushing forces exerted by the center mounting screw. In the stock design both of these elements are thrown together and you don't think about it. But when I started parsing things and modifying everything, it became clearer and clearer that both elements are fighting each other, because it depends on how much clearance I need under the spider (stock design simply overshoots this immensely and doesn't worry about it, and accepts the carpet of grounds left behind as a result.... something I won't accept). So, when juggling the spider sweep height, I am also juggling the force holding the spider from spinning on its own (slipping) although I don't need much force in the spider... grounds aren't very hard to sweep to a hole. But really, at this point I am thinking of separating the 2 elements once and for all, which will require drilling a small hole in the bottom of the inner burr face, I think, and mouting a small pin, and making a small hole in the spider for that pin to drop into in order to act as a drive key for the spider. That way, I can stop worrying about driving the spider and concentrate on only the burr mounting now that I have a solid bushing under it.

walt_in_hawaii

Postby walt_in_hawaii » Aug 04, 2018, 8:51 pm

OR...
I could change the center bushing design again, and make it 2 part... a top and bottom, to clamp the spider in between? hmmmmm. That would mean I wouldn't have to drill the burr, something which would mean not being able to use stock parts anymore, immensely more difficult for the next owner who may not be able to drill his burr to match. And burrs wear, so it would need replacement eventually....

walt_in_hawaii

Postby walt_in_hawaii » Aug 05, 2018, 2:57 am

What was I thinking??
I pulled out the 1 1/2" aluminum and cut a new bushing out of aluminum:
Image
Its identical to the other one, just has a wider 'hood' on top, to lay over and grip the spider over a larger area. I had to recut the recess in the spider to match. I could have reused the delrin sleeve, if I could have gotten it off... but it was on there too tight, so I just made another one and pressed it on. Next, I found an old package of 3/32" roll pins I had...
Image
then driled the spider with the new bushing in place:
Image
then I cut one of the roll pins about .200" with a dremel cutoff bit. The hole I drilled was .220" deep, because the spider is about .300" deep top to bottom. But I don't want the roll pin drifting down and scoring the grind chamber floor, so I made the hole blind ended (doesn't exit). I then used the vice to press the roll pin into the hole:
Image
I was so excited to reassemble the k10 that I put it all together and turned it on... and it didn't spin! :( .........
the spider just sat there, I was watching it through the port. WHAT???
glanced around and saw that in my rush I had forgotten to put the inner burr back on. Duh.
took the turret off again, torqued the inner burr back on, put the turret back on... and tested a couple handfuls of very lightly roasted ethiopians. Seems to work just fine! very clean. I will run this for a week and pull the turret off and report back, but retention should be almost nonexistent. Hopefully. Never can tell, especially when your workers are so careless they have extra parts lying around after final assembly.... :)

walt_in_hawaii

Postby walt_in_hawaii » Aug 05, 2018, 4:09 pm

The graveyard of dead parts :(
The spiders are pretty, but no use for them now since the grind chamber is a different diameter. The 8 legged one was never even used :(
All now in the trash...
Image

walt_in_hawaii

Postby walt_in_hawaii » Aug 11, 2018, 4:19 am

OK, so its been just about a week with the new modified grind chamber. Took it apart tonight to check on it (I pull about a half dozen singles a day, give or take a couple), all pictures are as-found, no special prep, no vacuum, this is how it is in the raw state after normal use, mostly just brush and a little bamboo skewer I use to reach in with. I don't seem to use the rocket blower much, it seems to throw things around a make it a bit messy. I do have the 'bopper' air accordion on the top which works really well. Top (outer) burr/turret:
Image
It looks about normal, same as a stock outer burr. But of course the real interest would be how much retention the grind chamber shows, as I already know the stock grind chamber accumulates a rather thick layer of grounds all along the back of the chamber, away from the port (see post #41 on page 5 of this post) where the brush can't reach the grounds. The new chamber:
Image
Pretty good! I'm stoked. There is a small amount of grounds on the top shelf of the grind chamber, where there is a small crack between the chamber and the top carrier... I think I could put in a small seal in this area, like a piece of rolled paper or something to block the grounds. Or just leave it, the space available is only about 20 thousandths of an inch or so, which can't matter much. Besides, the grounds that land here won't move, and won't get mixed in with any fresh incoming coffee so aren't really a concern.
But the grind chamber itself is absolutely clean. No leftover grounds at all in there... so, great, it will work just fine... for me. But, if that's as far as it goes, then this is just a simple exercise and doesn't help anyone else on HB. I think now that I know it will work pretty well, I will take the next step and start modifying the stock parts, because I now know I don't really need them.

walt_in_hawaii

Postby walt_in_hawaii » Aug 12, 2018, 3:41 pm

Now that I have a working grind chamber that operates with a confirmed zero retention, next thing to try would be to modify the existing spider to sweep more effectively; something that could easily be fitted to other k10s to correct the grounds retention problem.
Stock spider with its 3 fingers is put into a vice and each finger lopped off:
Image
Image
Then the bobbed spider is put into the holding fitting I made earlier, so it can be shaped:
Image
As before, the aluminum exhibits very strange properties. It is VERY sticky and does not want to machine cleanly. It leaves a lot of debris which weld themselves onto the cutting tools, even with lots of lubricant. Very frustrating, as the cutting must be stopped often and the cutting edges cleared manually. My fingernails are all nicked crazily (I used them to try to knock off the bits welded to the cutting bit). I took the spider out from the holding fixture, as I needed to reduce the bottom about .275" (about halfway) to make room for the delrin spider:
Image
During this process a rather large void is uncovered... this is a casting flaw in the original part, a bubble captured inside the part during the casting process from the factory. But it doesn't extend into the central axis, so should be fine. Notice this is not the only flaw; there are other, minor voids next to the big one, and along the flat portion as well. The delrin will cover this, so it doesn't matter. I next chucked up the large delrin piece to make the spider out of, and found that the largest piece I have isn't big enough... its 3" in diameter and the spider is about 3.550" in diameter. Ooops. Have to order larger delrin....