Odyssey Espresso (Argos Lever prototype) - Page 48

#471: Post by **Jonk** » August 26th, 2021, 7:52 pm

It's a 600ml boiler. Look at the graph, seems it'll start to run out of water before group temperature is an issue. Perhaps the plumb-in option will need some group management if trying to pull lots of shots in a row, but it looks a lot better than the average pressurized home lever that struggles after the second shot.

#472: Post by K7 » August 26th, 2021, 8:07 pm

Assuming you can get the average temp to the desired value with the smart controller (still to be proven IMO), I think the bigger challenge is to get consistent intra-shot temp profiles. GH and boiler water at 80C apart will have different intra-shot temp profile than when they are 40C apart, and 10C apart, etc. Temp decline will be steeper for higher delta. Could result in substantial difference in taste even if you nail the "average" temp.

#473: Post by **Diogenes** » August 28th, 2021, 9:33 am

Jeff wrote:I'm not Ross, and I can't say I completely understand the control mechanism, but at least as I think through it:

If you know the temperature of the water about to be introduced and the temperature of the group, you should be able to fit a model that predicts reasonably well what the extraction temperature will be. Based on that, you should be able to determine a boiler temperature for the current group temperature that gives you nn°C over the reasonable range of group temperatures. Use something to set that boiler temperature, and you should be able to avoid some of the scramble to pull at the right time as well as not needing to use special techniques to keep the extraction temperature from rising over a reasonable number of shots. Since you'll need something to do that calculation, might as well use a PI or PID controller to manage the boiler temperature.

Spot on...

Ross is an engineer in the aerospace industry. Running computational fluid dynamics (CFD) to these guys is like a carpenter swinging a hammer.
Modeling the Argos (or most espresso machines) is trivial for tools like Ansys Fluent.
So we don't know for sure but Ross most likely had the modeling complete before he tooled his first prototype and has been refining the model. Now he's pulling test shots and taking data, plotting across different temperature settings to compare the actuals with the model. At some point though... he's going to have to take data at sea level (he's in Colorado).

This is an example of a simple Ansys model.

August 28th, 2021, 12:33 pm

Diogenes wrote:Spot on...

Ross is an engineer in the aerospace industry. Running computational fluid dynamics (CFD) to these guys is like a carpenter swinging a hammer.
Modeling the Argos (or most espresso machines) is trivial for tools like Ansys Fluent.
So we don't know for sure but Ross most likely had the modeling complete before he tooled his first prototype and has been refining the model. Now he's pulling test shots and taking data, plotting across different temperature settings to compare the actuals with the model. At some point though... he's going to have to take data at sea level (he's in Colorado).

This is an example of a simple Ansys model.
video

I am actually wondering how many companies are using CAD based simulation to build espresso machines. I totally believe Ross did, but the big home espresso machine companies surly don't. For a lot of applications it surely is not necessary, you can do those calculations the old fashioned way or just don't do any when sticking an E61 on a boiler. I mean heating a block of metal is really not the typical use case for that kind of sophisticated software. Planes, cars, atomic plants etc. are the more typical application for software like that. Decent on the other hand might have used software like that.
Which other companies could be using software like that, especially in the home espresso market?

#475: Post by **Welshdog** » August 28th, 2021, 2:22 pm

coffeeOnTheBrain wrote: I am actually wondering how many companies are using CAD based simulation to build espresso machines. I totally believe Ross did, but the big home espresso machine companies surly don't. For a lot of applications it surely is not necessary, you can do those calculations the old fashioned way or just don't do any when sticking an E61 on a boiler. I mean heating a block of metal is really not the typical use case for that kind of sophisticated software. Planes, cars, atomic plants etc. are the more typical application for software like that. Decent on the other hand might have used software like that.
Which other companies could be using software like that, especially in the home espresso market?

Maybe they should use it? Just because you can slap an E61 on a boiler and get something that works, doesn't mean it could not or would not be improved by doing some analysis. Maybe that kind of research is exactly what these companies should be doing if they want to improve their product?

#476: Post by **misprint** » August 29th, 2021, 2:57 pm

Anyone know if there is a cut off for the first production run?

#477: Post by **Seoulenya** » August 29th, 2021, 3:36 pm

GregoryJ said "...he said he's not planning to cut off preorders." on an August 19th post.

Also, if all the orders since #70 have included an Argos, we're in the clear. I'm order 274.

#478: Post by **nguye569** » August 29th, 2021, 4:23 pm

He posted on IG 3 days ago that we've hit 200, so we're in the clear for production to start.

#479: Post by **bgn** » August 29th, 2021, 5:16 pm

I'm tempted by this. But still have a question about piston seals. I would imagine that they will be easy for the users to service the piston and replace seals, but still I'd like to know if they are custom or commonly found and if they are silicon.

August 29th, 2021, 7:24 pm

This is what Ross posted about the seals:

"combo of off the shelf o-rings and u-cup seal. can all be found off the shelf i'll post up sizes for folks to source and replace from their local/ online store"
https://discordapp.com/channels/7269855 ... 7436271627