Radial Scrubbers

[wreckweasel]

Dwell time..... you'll have many happy hours reading up on all this

/Z

[Brian Garner]

Ok I understand dwell time or I think I do, thinking of engines here. But surely the lime reacts to the end point where no more co2 is absorbed. Or is the lime in (for example the inspiration) scrubber not totally spent when the usefull end of life is reached.
That last sentence should make sense. When there is no more active area on the lime then no C02 is absorbed. How does the dwell time change this? On a radial unit the gas path is shorter and gas travel speed is slower. On a tower then the gas path is longer and the speed quicker.
Thinking about a tower type scrubber there must always be a high concentration of Co2 at one end ( By high I mean realtively. Possibly 4%)
Even though this is shoved through the scrubber it is always replaced by a fresh expired breath and I assume the percentage Co2 is roughly the same. At any point in the dive I would expect to find the lime at the input to the scrubber to be in a high co2 mix. As the dive progresses the high co2 area moves into the scrubber as the lime is exhausted. As the scrubber nears the end of its life the majority of the scrubber is in the high Co2 state.
BUT the start of the scrubber is bathed in a High co2 mix constantly. Any lime in this area which is almost spent will still be able to absorb Co2 but not as efficiently as the fresh lime later in the scrubber. Does this not offset the longer dwell time.
To explain.
A grain of lime at the input end of the scrubber is bathed in a high Co2 mix. Though the co2 is constantly being pushed past the grain. A fresh co2 mix replaces it. THe gradient of co2 to the center of the lime particle must be constant. Again I assume the active area moves towards the center of the garain as the exterior is used. Does this not equate to a large dwell time.

Sorry if this is a bit long winded. Trying to explain my thinking.

Brian

[wreckweasel]

Brian. With radial or fat/flat axial scrubber, your gas velocity is reduced, so you get a longer time of contact with unreacted lime (dwell time).

This means theres less chance of CO2 breakthrough.

The problem with most axial designs is that the dwell time is reduced, so your chances of scrubbing are reduced. Something to bear in mind is that in axial scrubbers you tend to have a "band" of reaction, that gradually moves along the stack.

Theres some documents on mathematical modelling of CO2 absorption on the net somewhere, if you cant find them I'll put them online later.


/Zak

[Brian Garner]

If you can find theose docs I would like to read them. They may go a bit over my head but I will give them a look.
I appreciate what you say about the longer contact time in a radial unit. This must allow you to use the lime more efficently than a axial type scrubber.
Do you not get a band of reaction moving outwards in a center fed radial scrubber?

I realise that the radial is far more efficient. I just would like to understand why this is.

Many thanks

Brian

[apitkin]

This thread is getting interesting.

Thinking about it, I've just realised that dwell time in a radial scrubber is not uniform, because gas velocity will decrease with increasing distance from the centre. In a cylindrical centre-fed design gas velocity will be inversely proportional to the square of the distance from the centre. Therefore dwell time must increase with the square of the distance from the centre. So, a very short dwell time (but less lime) near the centre, and much longer dwell time in the greater mass of sodalime nearer the outside.

You could achieve the same effect in an axial scrubber if it was conical.

I think you do get a reaction front of sorts in a radial scrubber, but I suspect it's much more diffuse than an axial scrubber for those reasons. However, when I've used the indicator type of Sofnolime in my Prism, the granules near the centre are definitely the ones that appear used up first.

Thinking further, what about a spherical scrubber, with the gas fed into the centre? It'd be a bitch to fill though.

With any luck someone like Pete Readey who really does know about this stuff will be along to enlighten us.

Andy

[Rob Evans]

Originally Posted by apitkin

This thread is getting interesting.

Thinking about it, I've just realised that dwell time in a radial scrubber is not uniform, because gas velocity will decrease with increasing distance from the centre. In a cylindrical centre-fed design gas velocity will be inversely proportional to the square of the distance from the centre. Therefore dwell time must increase with the square of the distance from the centre. So, a very short dwell time (but less lime) near the centre, and much longer dwell time in the greater mass of sodalime nearer the outside.

You could achieve the same effect in an axial scrubber if it was conical.

I think you do get a reaction front of sorts in a radial scrubber, but I suspect it's much more diffuse than an axial scrubber for those reasons. However, when I've used the indicator type of Sofnolime in my Prism, the granules near the centre are definitely the ones that appear used up first.

Thinking further, what about a spherical scrubber, with the gas fed into the centre? It'd be a bitch to fill though.

With any luck someone like Pete Readey who really does know about this stuff will be along to enlighten us.

Andy

Oooo! Blatant plug for SMI there

Interesting though; those were my exact same thoughts on the matter. Given the above discussion, why do manufacturers stick with axial scrubbers?
Is it a cost of manufacturing issue or are they shying clear due to worries about sudden breakthrough when the scrubber's used up?
Is a CIS scrubber really as much of a bitch to fill as I'm told? How about the Prism?

[wreckweasel]

Scrubber design is a bit of a black art, thats why. You can change parameters linearly and then get bizarre results.

I harbour a secret desire to get a small workshop setup and start tinkering, but then I have to realise that Im fundamentally a lazy twat and would far rather enjoy some nice wine and cheese.

Maybe its a darwinian process....

[Drmike]

Originally Posted by Rob Evans

Oooo! Blatant plug for SMI there

why do manufacturers stick with axial scrubbers?
?

I believe radial scrubbers have a more sudden breakthrough when expired whereas axial tends to have a slower breakthrough. Also radial need more care packing. But I could be talking bollocs



There was a good artice in Nitrox diver mag end of last year - also talked about co2 measurement (IIRC thet were not supportive of the thermister method)

[ewan]

Originally Posted by Rob Evans

why do manufacturers stick with axial scrubbers?

Possibly because axial scrubbers are easier to make, neater and usually easier to fill and seal. When you actually test well-designed types of each back-to-back there's no significant difference.

Look at the independent tests on the Ouroborus and Inspiration - they both get close to 100% efficiency based on Molecular Products' figures and that's difficult to improve on.

[nigelH]

Originally Posted by ewan

Look at the independent tests on the Ouroborus and Inspiration - they both get close to 100% efficiency based on Molecular Products' figures and that's difficult to improve on.

One of the significant items must be that a scrubber is most heavily worked during the first part of a deep dive. This is when the gas flow is highest and the work load is up. The later part of the dive, hanging on a trapeze for hours, is low flow and the system will work at its best.

This will mean that in the real world good packing, to minimise channelling, is the key issue. What you want to do is minimise the amount of good lime left when the breakthrough becomes unacceptable. This is what also tells me that the Inspiration testing on high flow rates means that low flow rate me will have a lot of margin for error if I need to overstay my 3 hours dangling on a string staying with a buddy who is breathing a less than optimal OC backup mix.

What worries me is people making comparisons on quoted scrubber duration when one is based on 'worst case' testing and the other is 'average use'. I would bet I could put on the YBOD and sit here reading the web for 6 hours and still have lime to spare. It proves very little. A scrubber test needs to be at a lot of ambient bar and a deep water flow rate or it is just playing games.
*sigh*
Sorry. Testing things is a big part of my business. If the product doesn't work in the environment the customer is going to use it in I'm wasting my time.