oxygen concentrators

[Ahmdo Molah]

I have given up on my attempts to fix the DeVilbiss 525. Maybe some day I will try to replace the sieves.
All the years of using a semi-working concentrator ...].

Since you already have a working alternative, hence nothing to lose by giving your "DeVi" one more chance, consider the following*;

* Excerpt from DeVilbiss 525 Service Manual (p.15, COMPRESSOR Note #1 )
https://www.manualslib.com/manual/12...page=15#manual

" NOTE – A built-in thermal cutoff switch will shut the compressor off if it becomes overheated. This protects the compressor from damage caused by heat build-up. (Some models have an auxiliary thermostat mounted within the compressor compartment.)"

1. The "semi-working" of the compressor maybe due to the 'thermal cutoff switch' mentioned above, and the noise dampening foam maybe causing overheating of the compressor. I'd try removing as much of the internal foam as possible.
* My DeVi (515 series) is OLD so it does not have such fancy thermal sensor controlled throttle for the compressor, and when I first opened the housing the internal foam & adhesive were falling/crumbling/cooking on the compressor (which gets as hot as around 70 deg.C / 158 deg.F, clogging the exhaust fan underneath/compounding the thermal stress, and looked like a fudge/brownie batch exploded in an oven!

[ DeVilbiss 515 Foam Meltdown ]

DeVilbiss 515 Foam Meltdown.jpg

This explains why my DeVi took a little 'crap' of foam out of its lower exhausts(causing a brief nervous breakdown of the seller) when I was testing it on the day of purchase. I was more interested in the O2 analyser reading of 94% by 20 minutes of operation at 2~3lpm.

[ DeVilbiss 515 Foam Removed ]

DeVilbiss 515 Foam Removed .jpg
*The marshmallow-shaped compressor exhaust filter is temporarily removed in this photo.

2. Removing the noise dampening foam (ALL in my case, with the help of alcohol & some sticker remover spray for tough spots) would increase the internal 'air volume' (possibly cool air intake as well) around the compressor and may reduce the throttling of the compressor by the thermal sensor, allowing more consistent O2 supply.

The 'older' DeVilbiss units are built tough, but lack a heat exchanger & dedicated fan which most others (like your Nidek Nuvo 8) have. Even the newer DeVilbiss units only have a small aluminum heatsink block (10lpm unit) or an extended silicone hose connection from the compressor to the sieves/valves (5lpm units), which really are not enough considering how HOT the compressed air gets between the compressor and sieves.

My DIY (1m aluminum HVAC pipe) heat exchanger's intake brass coupling hovers around 65 deg.C/ 149 deg.F and its exhaust coupling hovers around 55 deg.C/ 131 deg.F, without a dedicated fan (only the beefy main fan under the compressor). The added moisture separator made of aluminum, and added brass nipples likely drop the temperature a bit further as well.

The noise level is not too bad, even with ALL the foam removed, with the front & back housing(covers) both on.

[Ahmdo Molah]

My previous post's mention of 'removal of ALL the internal foam' from a DeVilbiss 515 O2 concentrator necessitates an additional note specifically related to the exhaust vents of the concentrator's base.

DeVilbiss 515 Base

DeVilbiss 515 Base.jpg
*Note: the grey colored parts are my epoxy(JB-Weld) reinforcements for cracks in the caster pole sockets.

The main exhaust fan is mounted centrally (above the black rubber lined area), and exhaust air is blown down and split toward the left & right vents.

The top of the M-shaped exhaust airway is partially covered by the upper internal structure and partially by the base's foam cover, but when the base's foam cover is removed, the top of the exhaust vent airways get partially exposed.

DeVilbiss 515 Base [Partially Exposed Airways]

DeVilbiss 515 Base [Partially Exposed Airways].jpg


The exposed areas should be covered so that the hot exhaust can fully be directed out the vents. I considered using various foam materials treated with flame-retardants but considering their tendency to out-gas nasty chemical fumes, I opted for felt(natural wool)* as it has flame-retardant properties, to make a cover.
* Thick felt sheets can be expensive, but felt(sheep's wool) ironing mats are inexpensive and easy to get online.

The original base foam is attached to an identically outlined black rubber sheet, which served as a reference for my felt base cover.

DeVilbiss 515 Base [Felt (Wool) Cover]

DeVilbiss 515 Base [Felt (Wool) Cover].jpg


DeVilbiss 515 Base [Felt (Wool) Cover & Original Foam Cover's Rubber Sheet]

DeVilbiss 515 Base [Felt (Wool) Cover & Original Foam Cover's Rubber Sheet].jpg


Cheers.

[Doug Fattic]

I have given up on my attempts to fix the DeVilbiss 525. Maybe some day I will try to replace the sieves.
All the years of using a semi-working concentrator made so frustrated that I bought a brand new Nuvo 8, which as the name suggests produces 8 lpm of oxygen.
Oh, the wonders! It has made fillet brazing enjoyable again. On the other hand, I have learned that I don't really need 8 liters of oxygen.
The flame is so much hotter with the new unit (probably due to better oxygen purity) that for most work 3lpm is enough and my torch definitely can't handle more than 5lpm.

Hmm, last week I sent Yuriy money raised by my cycling friends to replace the tools stolen by the Russians after they broke down the door of our workshop in Bucha. They took general tools like our vises and drills and metric wrenches but not the bike specific equipment (as far as I know from Yuriy's report). Fortunately Yuiry put the Anvil fixtures in the steel shed next door that the Russians couldn't mange to break into. In the other storage shed they did manage to break in but just left my road bike. It was still there with the door wide open several weeks later when college people returned.

What i'm wondering out load is if getting an oxygen concentrator would be a smart purchase and within our replacement tools budget. That huge Oxygen tank we use took 2 people to move when it needs to be refilled. I also wonder if the oxygen supplier is still in existence? I'm not sure where Yuiry went to get it refilled. Of course for all the reasons I promote I would like to switch over anyway so the oxygen supply is more in our control. That modest workshop is where I started using propane instead of acetylene. I've been going through old pictures making frames in Ukraine for presentations I have to give this week end at a cycling event.

[Arttu]

I don't have a heated (or insulated) shop at the moment and need to do a bit of brazing. DeVilbiss states that the lowest operating temperature is 5°C /41°F, will I risk damaging the concentrator if I run it close to, or below freezing? Or am I better off heating the workshop to above the recommended temperature limit with a space heater?

[Doug Fattic]

Arttu, I don't have enough knowledge to answer your question of how cold is okay to run a DeVilbiss concentrator (besides what they recommend). When I was in Ukraine during the winter, I found it uncomfortable to work in the workshop unless it was at least 10º and preferably 15º C. I would turn on 2 space heaters to take the chill out. So I would think that for your own comfort, you would want to have your workshop heated to above the working temperature required of your concentrator.

One alternative would be to put the concentrator in a smaller space (like a closet) and use the space heater to heat up that smaller area instead of the entire workshop. One caution for doing that is that an oxygen concentrator needs a certain amount of space around it to function properly. It shouldn't be placed too close to the corner of 2 walls for example. Of course I don't have any idea of what the limits of surrounding space is required to function properly. Where are you located?

[Mark Kelly]

DeVilbiss states that the lowest operating temperature is 5°C /41°F, will I risk damaging the concentrator if I run it close to, or below freezing?

As I understand it, the problem is that the compressor causes moisture in the incoming air to condense* and the condensate can freeze in the lines and damage the unit.

Pre-drying the air is the usual solution if the unit can't be kept warm.

[bastia]

Hi everyone !
Thank to all for your questions and answers. It's a pleasure to benefit from a real experience sharing.

I've completed my oxy-propan brazing setup. It's the second time I test it.

I have problem in flame consistency. The cone is not stable, it vary in size and color rapidely. It is impossible to adjust it. I think that my oxygen flow change at each compressor piston stroke. Mayby my oxygen checkvalve (it's only a checkvalve) has a too strong craking pressure. But I'm not confident to run my setup without it.

Here are my options :
- special low cracking pressure checkvalve probably very expensive and hard to buy for a non professional
https://www.wittgas.com/products/gas...valve-ultra10/

- oxygen checkvalve for medical supply
https://www.linde-healthcare-element...ffnasenbrillen

I'm not sure that the second is realy a checkvalve. I have already installed a oxycon firesafe adapter, but it's not a checkvalve, the flow can go in the two ways, it's more a thermal fuse.

Did somebody have an advice ?

[Matthias]

Bastia,

I feel like you're barking up the wrong tree. It's not that I had particular experience that points to the contrary but there is several things in an oxy-fuel setup that can lead to pulsation and the oxygen concentrator might add one more. Meanwhile the checkvalve idea doesn't sound too convincing from personal experience aswell as reasoning. If your checkvalve actually seated during operation it would cut all flow in that instance and only crack open again once pressure on the supply side increased beyond the pressure that unseats it. This would mean your flame, if you could even sustain one under those conditions, would fluctuate very drastically, as in going from oxidizing to sooty and burning in ambient air back to oxidizing at high frequency. Perhaps show some pictures of the extreme ends of the ovserved fluctuation.
Now regarding my personal experience: I'm currently running 4 safety devices, two at the regs and two at the torch. Both are ancient, crusty and thanks to the labels being worn off mostly I have no idea what exactly two of the doodads are. They have simply always been on the setup and nothing catastrophical happened. They supply big torches for big flame cuts. But they don't cause any problems when being tasked with supplying the smallest torch and smallest flame I could find and adjust. With a reg set at below .1 bar they will still flow gas. So assuming at least one of the mystery safety devices contains a ball type check valve it's safe to say it opens at negligible pressure.
I noticed I too am getting a bit of fluctuation after switching to propane. I chalk that up to me using a new, cheap and no name propane reg and a short propane hose. There is a theory that a new regulator shows stronger hyteresis than a broken in one due to the surface finish on the plunger that is moved by the diaphragm getting burnished over time.
The shorter hose im using for propane with its decreased internal volume might do a worse job dampening the pulsations.
In my mind before making purchases, throwing money at problems to see what sticks, you could try narrowing it down. You could for example determine in an experiment, for each gas individually, the lowest pressure you can set at the reg and still get flow at the torch. If it's much lower than your operating pressures I see no way a check valves opening pressure could be the culprit. Now I don't want to be that guy telling you to do what is implied in the following. I'm just saying that catastrophic accidents are usually a result of user error, faulty equiptment, negligence and statistically occur only once per so and so many hours of operation. It follows that running the setup without safety devices for as long as it takes to witness a change in operation or its absence while maintaining proper protocols and being mindful (purge everything first, have a plan, make sure pressures are spot on) can be expected to be without bad consequence. Or you get unlucky.
Now say you ruled out the safety devices I'd suggest it's the concentrators output. You can always use a damper (large dead volume) in a system running compressible media. Like a longer hose, a bottle, water hammer damper ... anything rated for the pressure and pure oxygen.
Then theres the thing with expensive regs, as in larger diameter diaphragms, two stage etc. I have no experience regarding those but judging by how good things normally are with standard regs I am sure those will only bring about marginal gains and wont make the difference between unusably bad fluctuations and a working setup.
I hope you can find out whats up and stay safe in the process.

Best regards,

Matthias

[bastia]

Thank you a lot for your answer. You were right ! You saved me from wandering away.
I didn't suspect the propane regulator, the open flame without oxygen seemed relatively stable.
But I tested without the checkvalve, it changed nothing. I added 12 foot of oxygen hose with the same outcome.
I added more than 12 foot of propane hose, it might have changed something, but not significantly.
I increased (lightly, switching from 3 to 5-6 psi) the propane pressure, it changed significantly the stability of the flame. It's now a lot more stable. It's not perfect, I still have little variation in the cone size (the cone can vary from 0,2 inch) , I don't know if it's normal or not, but i think variation light enough to let the setup usable.
I tried to film, but this variation need a good camera to bee see. I would try again.
My propane regulator is designed to operate propane gas at low pressure (up to 21 psi). It's not a cheap (or a fancy double stage) one, but it come from a well know and reliable local (french) manufacturer (Le lorrain).
Again, thanks !

[jimgracks]

Thank you a lot for your answer. You were right ! You saved me from wandering away.
I didn't suspect the propane regulator, the open flame without oxygen seemed relatively stable.
But I tested without the checkvalve, it changed nothing. I added 12 foot of oxygen hose with the same outcome.
I added more than 12 foot of propane hose, it might have changed something, but not significantly.
I increased (lightly, switching from 3 to 5-6 psi) the propane pressure, it changed significantly the stability of the flame. It's now a lot more stable. It's not perfect, I still have little variation in the cone size (the cone can vary from 0,2 inch) , I don't know if it's normal or not, but i think variation light enough to let the setup usable.
I tried to film, but this variation need a good camera to bee see. I would try again.
My propane regulator is designed to operate propane gas at low pressure (up to 21 psi). It's not a cheap (or a fancy double stage) one, but it come from a well know and reliable local (french) manufacturer (Le lorrain).
Again, thanks !

What is the brand/model of your oxygen concentrator? Some of them (usually the smaller portable machines) only deliver O2 in pulses vs. a continuous stream. Check inside your machine to see if it has a holding tank. If not, you may benefit from installing a small pressure tank in line between the machine and your hose to balance out the flow.

-Jim G

[bastia]

I have a visionair 5 from airsep.
I did not open it (at the moment). The schematics of the visionair 2 and 3 indicate a mixing (?) tank / a assy (?) tank and in the two cases a regulator. I think that my oxycon have a (little) tank too.
Add a tank is a good idea to test/solve the problem, but technically I wouldn't really know how to add a tank (which oxygen container to use).

[Doug Fattic]

bastia, i'm wondering if you can put in-between your concentrator and your torch handle (on your oxygen hose) an air carry tank. These are small (mine is 7 gallons) tanks designed to get filled with pressurized air from a compressor and take it somewhere to fill up a flat tire (or whatever). I wouldn't use typical quick release connectors to connect the output hose to the tank (these QR connectors are used in almost all air tools so they can easily be connected and disconnected from the air line) because they require more pressure to operate properly than a concentrator can provide. I'd just connect the oxygen air line directly into and out of the tank. This I would assume help smooth the air output flow. The disadvantage of course is that you would have to run the concentrator long enough to purge the tank of ordinary air until it became full of only oxygen.

As I understand it (maybe) a concentehtrator has 2 pistons (or whatever they are) working together. One is on while the other is off and repeat. This is what causes the pulsing as they go back and forth. I think my Devilbiss has a "bladder" that is like a tank that smooths out the air flow between pulses before the oxygen leaves the unit.

Here is my air carry tank.
IMG_7662.jpg

[Mark Kelly]

As I understand it (maybe) a concentehtrator has 2 pistons (or whatever they are) working together. One is on while the other is off and repeat.

Usually called beds. Each bed is a pressurisable container full of a molecular sieve, usually a zeolite*.

Each bed is charged with pressurised air, the zeolite adsorbs the nitrogen and the oxygen rich remnant gas is passed to the output. The bed is then depressurised which desorbs the nitrogen from the zeolite, this is exhausted to atmosphere then the process starts again.

Common machines have two beds so one is charging while the other is discharging. Some machines have multiple beds to reduce the output ripple.

* Zeolites can be thought of a spongy minerals with a huge number of incredibly tiny pores. The chemistry of the zeolite can be altered so that these pores become selective for particular molecules, hence the name molecular sieve.

[bastia]

Thank you for this advice. Maybe a 7 gallons is a little big, and it would work with a truck brake air tank (3,5 gallons).
I'm going to continue my tests, and if the variations persist and are blocking, I'll try this solution.

[Doug Fattic]

Thank you for this advice. Maybe a 7 gallons is a little big, and it would work with a truck brake air tank (3,5 gallons).
I'm going to continue my tests, and if the variations persist and are blocking, I'll try this solution.

Yes I would try and find a much smaller tank. I just had that 7 gal tank handy as an example so you would get the idea of what I was trying to explain. A smaller unit would take less time for oxygen from the concentrator to purge the air and replace it with almost pure oxygen. Concentrators if working properly get over 90% pure oxygen but not 100%.

[Mark Kelly]

Concentrators if working properly get over 90% pure oxygen but not 100%.

My understanding is that the common nitrogen removal zeolites eg X13 are not selective for Argon, so it remains in the output gas. Since air is about 21% Oxygen and 0.9% Argon, that puts a limit on the purity achievable.