Ameen's designing efforts

For Short Term Learning Discussions ONLY. This area is for CURSORY questions and connecting with other users ONLY. ALL technical contributions need to be made in the appropriate forums and NOT HERE. All posts are temporary and will be deleted within weeks or months. You should have already search the extensive FAQs in each of the forums before posting here as your question may already be answered.
Post Reply
Ameen Aydan
Posts: 116
Joined: Sun Apr 15, 2018 3:33 pm
Real name: Ameen Aydan

Ameen's designing efforts

Post by Ameen Aydan »

Hello,

I have recently finished my design and capital evaluation on my fusor. Bellow I have attached an IGS file and the link to my google sheets document.

Now for some self reflection. For the Feed through I found a listing on eBay at about $500. I'm trying to bargain down the price at the moment, but if that is unsuccessful, I'll just use a compression port fitting. I can get to that when the issue is present. For my gas system, i'm using a sapphire variable leak valve. I know, I know, "Ameen that costs way too much not to mention complicated! There is no way in hell you can use one". A simple search on eBay resulted in multiple listing with the ol' "I don't know much about this stuff" and a reasonable price. It also turns out they're much easier to use than the normal deuterium system. I can talk about it more if your interested. View ports are readily available on eBay so that's not a problem. We all know about the MKS 901p so that doesn't need much explanation.

As for the cost, i'm not sure how i'm doing so i'm reaching out to this forum in an attempt to see how efficient my costs really are. Is there any way I can further decrease the price? Is this the best I can get in terms of cost? Any feed back is highly appreciated.

I will be designing the full gas and vacuum pump system as well as the cooling system. I'm thinking of using the Peltier device from free old computers as my cooling system.

Any criticism is HIGHLY appreciated. I will post update as I progress through my designs. Once I begin to receive Item (hopefully in December), I'll display my efforts.

IGS file and excel cost document can be found here: https://drive.google.com/open?id=1ir0Cl ... nOBDUNkhwd

Ameen Aydan
Michael Bretti
Posts: 175
Joined: Tue Aug 01, 2017 12:58 pm
Real name: Michael Bretti

Re: Ameen's designing efforts

Post by Michael Bretti »

In terms of cost, I think your overall prices are pretty reasonable. A 4.5" 6-way cross for $395 total including shipping is a decent deal, usually 6-way cross are quite expensive even used. As always, negotiating for the lowest price is always preferable if possible. I would say that for your zero-clearance reducers, you may be able to find them a bit lower - I found my 6" zero clearance reducer for a bit cheaper on eBay in perfect condition, but it depends on your luck of what is available at the time you are searching. For shipping from LDS, I would expect about $15-25 as well.

In terms of your chamber design, I think it is great they you have taken the time to plan and design it out, and always encourage everyone I talk to to fully plan out your system before you start buying parts. Looking at the chamber cad file, it is about as simple as you can get while having all of the necessary inputs (vacuum gauge, gas, feedthrough, viewport) as well as a good amount of internal space and an extra port for diagnostics if needed in the future. I think your vacuum chamber design will serve you excellently going forward for your plans with a fusor.

In terms of cooling, I should share some of my experience, and particularly some cautions with peltier systems. If you are using a turbopump, then this stuff is not an issue as opposed to a diffusion pump. Right now I am using an 850W, 600 L/s diffusion pump. I decided to go the route of a closed-loop peltier chilled system to keep everything compact and contained within my test stand. I am running x8 TEC1-12708 peltiers on a large custom built copper chiller block with full thermal insulation, as well as two coolant tanks - one for the diff pump cooling, and one for the cooling of the peltier hot side. I also have several large heat exchangers for both loops, and am running the peltier module in parallel with the diff pump loop, for a triple-loop system. The entire system was built from scratch, and it has been a massively challenging effort. Peltiers work great just out in air with some hot side cooling and no load - they will get super cold very fast. However, as soon as you add any sort of thermal load, system optimization becomes critical. I haven't been able to test it with the full diff pump load yet, rather with dummy heat loads, but it has been a major challenge, as well as a significant investment of time and money. The system is also heavily instrumented with 6 custom thermocouple probes, as well as flow monitors for each loop and PWM speed control for the pumps. You can see the full build details here:

http://appliedionsystems.com/portfolio/ ... p-cooling/

As well as the peltier chiller block build here:

http://appliedionsystems.com/portfolio/ ... er-module/

Finally, the first round of testing (this was a while ago, before I discovered the fans on my heat exchangers were largely useless, and needed replacing):

http://appliedionsystems.com/preliminar ... ng-system/

In order to anticipate the thermal gradients and loads encountered with the system, I also performed an in-depth thermal analysis of the entire diffusion pump and baffle assembly, which was the result of numerous iterations and many, many hours of cad and research:

http://appliedionsystems.com/portfolio/ ... -assembly/

Going back to your system, what type of pump are you running? Is it a diffusion pump, and if so, how many watts? Finally, what is your end goal temperature, as well as the recommended operation temperature of you pump (and baffle if you are using one)? While a peltier chilled system can work, it is a massive effort to optimize and get working well. Mine has already gone through multiple design changes, and is still in the process of optimizing. I would caution you that if you are trying to manage close to a kW of thermal load and regulate the temperature down below ambient with a peltier chiller in a closed-loop system, it will be a challenge. While you can run diffusion pumps warmer than optimal without much issue, you will generally find that diffusion pumps are specified for optimal performance between 15C and 25C cooling temp. The baffles are also more effective when they are run cooler. They will still reduce backstreaming running higher than optimal, but if they get too warm they can exhibit oil "creep" along the surfaces, and are also generally optimal at lower than ambient, closer to 15C if possible. Using higher quality oil such as DC-705 equivalent or better will help reduce these issues, although at a higher cost.

As you can see from my cost analysis of the system, it was not cheap as well. While I learned a lot, some mistakes were quite costly, and could have been put to better use for my vacuum systems. If your system has open loop cooling, that relaxes a lot of cost and challenge. Closed-loop is ideal if you want a self-contained system, but thermal management adds to cost and complexity. Chances are, you don't need to run a diffusion pump at ideal or optimal settings for your purposes, and only need to provide cooling for safe operation of the pump. A smaller pump however would be easier to manage, and my system is rated to also run the larger 1450W, 1200 L/s diff pump I have in the future if I wanted to upgrade.

In summary for cooling, I would first evaluate your goals and system requirements. What type of pump do you have? How many watts of thermal dissipation is needed? What is the optimal coolant temperature range? Baffle specs? Do you need a closed-loop system, or is open loop fine? What advantages to peltiers offer over other methods of cooling, as well as their disadvantages? Is your system designed to be compact and self-contained, or can you have external auxiliaries for cooling? I wrote an article a while ago looking at my mistakes and lessons learned in my first cooling system redesign that addresses some of these issues: http://appliedionsystems.com/learning-f ... -redesign/. As a note, the article is old and the thermal models referenced in it no longer apply, as my design has undergone massive changes since then, but still gives a good overview of lessons learned in system redesign. Even now, I am still working on some bugs, and will have to write a new updated article on these changes in the future. Is this level of cooling control necessary for a typical amateur fusor? No, not at all, and most others run much simpler and just as effective setups. However, it depends ultimately on your end goals - I am not working on a fusor, and my system requires tighter control and feedback.

Your chamber design looks very good, and seems like you are on the right track. One thing to also note, that it may be a challenge to adapt your pump to the chamber if the pump does not have a CF flange, and may require a custom plate or several adapter plates, like in the case of my system. Getting mismatched flanges and baffles to fit together with a chamber can be challenging, and may require some creative design.
Last edited by Michael Bretti on Fri Nov 16, 2018 11:39 am, edited 1 time in total.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Ameen's designing efforts

Post by Richard Hull »

The 6 way 4.5-inch cross was a good deal. Now to locate (6) 4.5-inch blanks or adapters for vacuum, viewing, HV feedthrough, etc. Did the cross come with any conflat blanks or adapters?

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Ameen Aydan
Posts: 116
Joined: Sun Apr 15, 2018 3:33 pm
Real name: Ameen Aydan

Re: Ameen's designing efforts

Post by Ameen Aydan »

Hello,

Though I have not purchased any of the parts yet, I'm going to start in two weeks time. I just wanted to remind you that these prices are in Canadian so converting them will display even better prices.

As for the parts, I have already located them with the links. I just didn't want to show the links for obvious reasons.

I just offered $150 USD on a TCU 050 preffier turbomolecular pump and it was accepted. He didn't say what the condition was in but it was put as "used" and not "for parts". So incase something goes wrong I can return the pump. The pumps has a 4.5" conflat flange as it's connecting port.

The controller is going to be a bit difficult to find at a good price, but nevertheless, I can find it. If that doesn't work out I know some people on this forum have made a DIY controller so I might just consider that. It's gonna be hard, but I'm here for the challenge.

I must say, your closed loop cooling system is a true master piece. I would love to have something like yours but for now I will try to keep it simple. If the Peltier cooling system is going to take that much effort, I might postpone it for the future when I finally achieve fusion. But in all seriousness, I have never seen such a well designed and thought out system. I truly respect and appluad your efforts, Michael Bretti.

Ameen Aydan
Michael Bretti
Posts: 175
Joined: Tue Aug 01, 2017 12:58 pm
Real name: Michael Bretti

Re: Ameen's designing efforts

Post by Michael Bretti »

Ameen,

If that turbopump is in good, working condition, then you are pretty much golden. Having a pump that can directly mate to your 4.5" CF chamber is a massive boost and makes things much easier. No need to mess around with custom adapter plates and o-ring seals. As long as the seller allows you to return the pump for a full refund if it is in bad condition, then it won't be much of a loss. Be aware of the risks though that the pump at this point could be anywhere from completely useless to lightly used. It's a gamble, but if you can return it at no charge then it's worth a shot taking a look at for that price.

Thank you for your compliments. I wouldn't say that the cooling system is a masterpiece yet really, and it's honestly given me more headache than anything else so far. Design is one thing on paper, but the true qualifier is testing it during full system operation. Still have to see how it performs with the full diffusion pump load. If you are running a turbopump, then cooling will be much easier to handle than a diffusion pump. Some turbos offer options for either air or water cooling, with different performance specs for each mode. I wouldn't expect however that water-cooling a turbo would be as intensive. For now though, if you are on a tight budget and just want a working system to begin exploring the fusor, simple is best. You can always design and plan to upgrade in the future if you want. If you work with the system ahead of time and know how it behaves (like how hot the turbopump gets during operation), then designing a more sophisticated water-cooled closed loop system will be easier going forward in the future.

My goal has been to fully engineer every aspect of my system, know all expected parameters and estimated performance, and provide all of the engineering resources to other makers. If you are interested in further more intensive design for high vacuum systems in particular, I am currently working on some exciting simulations for pumping and molecular flow for my chambers. I will be posting the results when I finish, as well as links to the project pages within the next few weeks. Like all of my other listed project and simulations, the analysis, explanations, and input files will be freely available. So far the simulations look like they agree with my initial hand calculations that I have posted earlier, and verify the validity of the vacuum system design going forward. Once these simulations are done, then I can finally start the real design work on my major experiments. I will also be using these results to simulate ultimate pressures during operation of my systems with actual inputs from things like propulsion engines/beam loading (depending on the system and experiment run).

Although I am still saving up to finish my system and get it running, I have been able to still design and "run" my systems virtually to know what to expect. CAD, thermal, sputtering, and now molecular flow, I can perform virtual tests and qualify estimated performance, even without money. It takes an incredible amount of time and effort, but is well worth it in the end. All with freely available resources too - it is amazing the level of engineering design and analysis you can do with what is freely available. However, it all starts with solid background research.

I also applaud your efforts for taking the time to actually plan out and budget your system ahead of time, and doing the research in advance. For such a complex and expensive hobby, it is amazing how few people actually starting out do this, and instead try to dive right in without taking the time to do any planning or research. Your chances of success will be much higher, you will better optimize your time and money, and in the end, you will learn a lot more and have a much better system than someone who just rushes to cobble together a system without any planning or thought.
Post Reply

Return to “New User Chat Area”