Visit the forum instructions to learn how to post to the forum, enable email notifications, subscribe to a category to receive emails when there are new discussions (like a mailing list), bookmark discussions and to see other tips to get the most out of our forum!
Alternate CEB press designs: mutual improvements
  • 76 Comments sorted by
  • One thing to keep in mind is with the current setup of the CEB Press, you do not stack the bricks in the orientation of compression because there is too much variability of that height. The main reason for this is that ram is controlled by positioned magnetic sensors which trigger a timing sequence to try and ensure a desired compression pressure. When you lay them you orient them so that direction of compression is oriented inwards and outwards vs up and down on the wall. This keeps the small changes from affecting the alignment of the wall.

    The easy fix to this is to have the compression chamber limits to be set by hard switches. The downfall is you may not get the desired compression pressure if the soil mixture changes. But ultimately the hard switches are how most other industry standard CEB Presses work to produce consistent brick.
  • Vote Up0Vote Down February 2012
    Yeah, look it's pretty evident that there has to be quite a bit of quality control of the soil mixture when working with interlocking bricks. But it's probably not much more than what a building engineer would want to see as standard practice when using their non-interlocking cousins.

    I wonder Mike, how hard is it to calibrate a press to get the desired brick size for a particular soil mixture? I guess what I'm trying to say is, you would think that the harder a brick is pressed, the stronger it would be? If that is the case, you would want to be pressing bricks to almost the capacity of the machine? And, given that what if a particular soil mixture requires just a little bit more oomph to get it to the correct height?

    As far as a production line, I'd be mixing a batch of soil (in a purpose built mixer), pressing that batch of bricks. Test 1 brick per batch. If that brick fails QA the entire batch is sidelined (I'd still use them, but maybe for garden beds or some other non critical function).

    As far as using the Liberator, it "might" be possible to shoe horn it into making interlocking bricks, but there's just too many things about the design I don't like that turns me off it.
    e.g. It would seem you'd have to take large parts of the machine apart, just to change a brick mould.

    Chuck, I think you forget that the area that you've put the rails, there's massive draw that fills that void, operating back and forth. So your rails would have to be inside the draw itself, but the brick in it's entirety fits in the draw's ejection chamber. Unless I've missed something, there really would have to be a significant rethink in the Liberators design to incorporate interlocking bricks, and given it's complexity of design (compared to the examples I had links to) I really couldn't be bothered. I'd prefer to start again as I can see, straight from their operation, how those machines work.

  • Vote Up0Vote Down February 2012
    p.s Here's what I've got so far : It's not much, but shows the press box and press plate.

    CEB.png 18K
  • Is the desire for interlocking blocks based on a comparison to non-interlocking blocks in use, or is it just a preference? I saw some examples of interlocking blocks an it looked like they were being used with rebar and cement/mortar (at least down channels through the blocks). Seems like if that's necessary they can't be all that superior to non-interlocking blocks. A video of that guy building the FabLab with CEB's had a conversation between him and Marcin where the guy seemed to think it was obvious that cement would be poured down the hollow columns after they were finished. From a systems point of view, it seems like there would have to be an impressive performance benefit to interlocking blocks before it would be worth tooling up for them, given the increase in machine complexity required to press and extract them. Just curious if anyone has done some analysis on any performance differences.
  • Vote Up0Vote Down February 2012
    Hey Matt,

    First up, the interlocking bricks that are designed to have rebar and concrete poured into them I'm not interested in. I think they're used for a replacement for high rise construction, I'm not really sure. The interlocking bricks that I want to press are basically a solid brick with interlocking parts. The benefits to interlocking bricks aren't really in the performance of the brick itself but in labour saving.

    Interlocking bricks can be laid with very minimal (almost no) motar, and given that with the correct soil type CEB bricks in general can be laid almost as soon as they're pressed, you can pretty much build the wall as the bricks are pressed. The interlocking means that unskilled labor can be used, the lack of motor means there's minimal effort required to line the bricks up.

  • I thought CEBs in general were laid without mortar. Like, all you had to do was keep them wet and they'd bond or something.

    I should preface this thought by pointing out that I have no experience in seems like the reason laying blocks would require skill is that they come out of the press slightly different shapes. If a machine is accurate enough to make interlocking blocks precisely enough that they always fit together correctly, then wouldn't it be precise enough to make rectangular blocks that always fit together correctly?
  • Vote Up0Vote Down February 2012
    Hi Mat, let me say I'm no expert either, but from all the literature I've read, this is my understanding:

    Standard CEBs require a mortor course. It is this mortor which creates the bond between coarses. I doubt just wetting the CEBs would allow them to bond together given that water penetration is one of the QA tests for CEBs. Interlocking bricks don't need the motor because the bond between coarses is engineered into the brick.

    Provided that standard CEBs require a mortor coarse, there is a bit of stuffing around with the mortor to line the brick up and basically squish it to the desired depth. String lines are used to accurately run the depth along the coarse. So regardless of if the brick is a standard size or not, it is the required use of the thick layer of mortor which makes the job of laying them slower. The same applies to standard bricks, and the interlocking bricks you can buy from brick suppliers (used for garden beds and retaining walls).

    The way I see it Matt, it boils down to this: Does getting the tolerances down in the machine design more difficult? Yes, but you do your thinking once, rather than every time you lay a brick.
  • So, is OSE making a mistake laying their CEBs without mortar? Do you still need mortar between each coarse if you, like, stucco the wall after it's built or something like that?

  • Vote Up0Vote Down February 2012
    Hey Matt, I thought I saw them use mortor in the construction vids at FeF?

    But yeah look, as I said I'm no expert. I've taken most of my knowledge from a commercial Aussie site : and it would seam that the use of mortor is expected. However, I did find this building code : which states that dry laying (stacking) is possible, provided that: adhesion is "proved", the wall thickness is not less than 10 inches (254 mm), and a stucco mortor or plaster is used. So you might be onto something there.

    I'm not sure what they mean by adhesion is "proved" though, do you say hit it with a sledge hammer?

    I'm going to continue with my design in any case because as I said, there's quite a few aspects to the Liberator I don't like, (mainly access to the pressbox to allow for swapping out moulds). I think I'd prefer to build a wall whereby the mortor isn't intergral to its structural integrity, mainly because I'm a lazy bastard and would build the wall and 2 years later get around to plastering it.

  • Oh I hear ya. The Liberator is just the first generation.

    "CEB construction, unlike standard masonry, does not require portland cement based mortars or thick mortar joints (1/4-1/2"), although we have seen CEB walls built this way.  Earth block builders have dry stacked walls, stuck them together with a thin slurry made from the same material as the blocks, and mixed thicker mortars of clay and sand.  We have found that the slurry method works well for unstabilized blocks because it penetrates the surface of the blocks and sticks them together."
    "Slip mortars.  Liquid mud slip mortar is allowed, providing it is made of a compatible soil that is screened to eliminate aggregate
    larger than one-eighth (1/8) inch in diameter.  Water may be substituted for slip or other mortars, providing adequate adhesion
    is demonstrated.
    C.Stacking.  “Dry stacking” of compressed earth blocks is allowed providing that adequate adhesion is demonstrated, the wall is to
    be stuccoed or plastered and the wall is not less than ten (10) inches in thickness."

    "The interlocking shapes of these improved bricks can help to reduce the skill level needed for homeowners to build their own homes. In addition, several layers of blocks can be placed in the wall at a time. A supervisor can then check that the wall is straight and plumb before giving the OK for mortar to be poured inside the hollow blocks to lock them together, thus eliminating the small openings between blocks which would otherwise allow for air infiltration."

    Seems to have a lot to do with the particular characteristics of the block and the context it's used in. At any rate, I can't see how interlocking blocks would hurt assuming the machine isn't significantly more complicated or slower.
  • Vote Up0Vote Down February 2012
    Yeah, I guess that's my thought, however I've just been reading this : which doesn't fill me with a huge amount of confidence when they say

    "We also found that
    it was difficult to nearly impossible to make blocks that were all the same
    height with the vertical press."

    Looking through the Cons (, a few that stand out (from the irrelevant) are : The larger press required, The larger structure and Irregular block thickness (height).

    With any CEB press there are 4 variables to make the brick: length, height, depth and pressure. To get constants in any of these, the others have to be variable. Which means to make an interlocking brick with the 3 dimensions constant, then the only thing that can vary is the pressure.

    Which brings me back to my earlier quandary: If a depth control is added, is the pressure difference to acheive a uniform brick feasible?

    I'm starting to think that it's all too hard.

  • I don't know that those sound hard to balance. If you "over compress" a brick it's just a little stronger than it needs to be. So, you need two sensors on the machine. One sensor would stop the compression when the brick reached the correct dimensions; the other sensor would stop the compression if the brick hadn't yet reached the correct dimensions, but the pressure in the machine was building up too high. And that first one you might be able to accomplish with the structure of the mechanism, rather than a sensor. So really it would only be one sensor as an E-stop in case there happens to be too much dirt in the mold and it just won't compress to the dimensions you want.

    Is it possible to compress a brick so far that it becomes weaker?
  • Vote Up0Vote Down February 2012
    Is it possible to compress a brick so far that it becomes weaker?
    I doubt it, with pressure comes heat, soil + pressure + heat = rock.

    Yeah I was thinking that too Mat, basically have a mechanical adjustable depth stop, so that when the press foot hits that the machine would try to compress until the high pressure signal fires.

    I've got a gut feeling that the pressure to depth ratio is exponential. e.g The depth would reach a certain point whereby to get a depth reduction ever increasing amounts of pressure are required. I could be wrong, it would be good if I was.

    At this point we need some data, and or field testing to see what pressure is required to acheive an acceptable brick, what pressure variations are required to acheive a uniform brick given standard soil variation.

    As a side point, I found it interesting that the guys at adobemachine say that an advantage of their horizontal machines is that "
    Can make a decent block from raw moist soil in most
    cases" yet they also sell soil mixers?

  • Vote Up0Vote Down February 2012
    Wow, this is totally the stuff that shot through my mind in the solution of this problem when I looked at the press under construction last summer and saw how they were trying to control it.

    Let me go over what I was thinking for a mainly software solution.  You add a couple of sensors - a reasonably high sample rate digital hydraulic pressure sensor on the main ram circuit, and a ram position sensor such as is used on cnc machining devices (high precision short latency contaminate shielded).  you also add a 'quality kicker' - any brick that doesn't meet qualifications gets kicked out of line, marked, whatever to make sure its known it is out of quality spec.

    You set in software the position at which the brick is the right size
    You set in software the position at which the brick is ready for dirt
    You set in software the position at which the brick is ready for eject
    You set in software the slack available in the tray clearance for ejecting oversize bricks.
    You set in software the threshold of compression pressure required for a good brick.
    You set in software the threshold of compression pressure that is too much.

    So if it makes a brick and the pressure is in the green zone, all okay.
    If it makes a brick and the pressure is in the too-low zone, actuate the kicker after ejection
    If it makes a brick and the pressure is in the too-high zone, and there is clearance to eject it, eject and kick
    If it makes a brick and the pressure is in the too-high zone and there is no clearance to eject it, full stop and sound jam alarm.

    There is also the possibility that the full speed of the input hydraulics is too fast to make reliably precise bricks.  I would suggest making a backup circuit - once the main ram gets in the proximity of the end point, the main valve shuts down and a secondary, low-speed circuit does the position finishing.  Once the position is met, the main circuit can be used for the eject and retract cycle.

    That said, there is a bunch that can be done through mechanical means should you want to - adjustable threshold switches and actuators are actually pretty easy to get for hydraulics.  With appropriate configuration of the parts, you could replicate this behavior in a hardware-only controlled system.  

    But we've got a computer attached to the thing, for crying out loud, which is far easier to teach new tricks than any amount of welding and hydraulic lines will ever be.

  • Vote Up0Vote Down February 2012
    On the matter of interlocking bricks

    Its entirely possible to make the machine articulate to make interlocking bricks in many configurations - but it comes down to the fact that a simple machine like this simply can't do this without taking a series of compromises to articulate the faces and sides of the compression chamber - not impossible, but complicated.  This is the kind of stuff that they do in industrial molding machines for mass production of parts that is very precise, very complicated, and they get paid very well to do.

    Suffice it to say, its too complicated for this application.  Stick with simple bricks and we'll do better.

  • Vote Up0Vote Down
    February 2012
    would it be worthwhile to draw up some flow charts?
  • Vote Up0Vote Down
    February 2012
    I've been studying the plans for the CEB machine with great interest.  I am encouraged about the discussions here revolving around future design specifications.  I am especially interested in the topics of alternate design of the bricks themselves.

    I reside in Latin America and one of the building codes I frequently run into with CEBs is the need for vertical reinforcement (read 'rebar').  If the blocks can be pressed with two or three holes in them, building codes can be met.  It seems that more discussions revolve around interlocking bricks, and while that is a useful goal, it strikes me that a more important feature is that of the 'holes'.  This, as you are aware, I am sure, is a fairly frequently seen feature of the bricks produced by competitive machines.

    The one comment I saw from mrsquish regarding holes in the bricks concerned me.  He stated that he was not interested in holes because it lead him to think that necessarily meant higher vertical construction.  I intend to build one-story structures, but would like to be in compliance with earthquake codes (even in countries where they don't exist as such).

    The holes should be fairly small for several reasons:  integrity of the brick itself, the need for less mortar, speed of construction, and the preservation of the other inherent benefits of the CEB.  I would suggest that a hole diameter be not larger than necessary to insert two standard rebar rods

    I thought another voice might help decisions regarding feature prioritization.  I truly believe the lack of 'holed' bricks is a massive competitive disadvantage, especially if it can be designed in as part of the adaptable form pressing system.
  • From my experience working with CEB's the compression height really isn't a big deal unless they are grossly off. If you're dealing with fluctuations of less than a 1/4" you can easily make up for those. Keep in mind we're turning our bricks on their side so the height variance doesn't mess with the coarse level. All bricks are then 6" high, 12" long and roughly 6" wide. If there is variance, you just need to compensate by adding wider bricks to make up for a narrower brick. Believe me, I've seen these work even with wild variances.

    No, from all my testing you can't over compress a CEB. You'd just make sandstone or an equivalent. I haven't noticed any real difference compressing with 2000 PSI versus 3000 PSI. I got the same height using both pressures across 100 blocks of each. 

    Cement mortar = bad. It doesn't bond to the bricks very well (depending on the soil mix), and doesn't allow for shrinkage. It usually sets faster than a block dries, if using green blocks. Best suggestion is using a slurry made from the same soil as the blocks. Bonds amazing well, simply due to it being the same material. Can be used on green or cured blocks. I've seen a test wall done using that method and you couldn't break any of the blocks free using your bare hands. Actually took a saw to cut one out.

    As far as shaped blocks and the current CEB design. You could feasibly shape the 6"x6" ends  like <==<. Of course the V cut would have to be inline with the compression otherwise it would be sheared off when it ejects. For that same reason, you can't really shape any of the sides with keys. 

    Holes could be done, but you would have to add in another ram to retract the rods used to make the holes. These would have to be added to the press foot and retract down prior to block ejection. I can't see any other way of doing it with the current design. 

    In short, CEBs are robust and don't have to be precise in all dimensions if you have a decent soil mix. However, common sense should be used when building with them. In most cases you're likely going to put something over them and wont see the blocks themselves. 

    my .02

  • Vote Up0Vote Down
    February 2012
    is it feasibly to use a standard drill to cut holes in the brick after it comes out of the machine?
  • Some stuff I found:

    "A vertical press presses from the top or bottom of the block. if the block is 10 x 14 then the press area is on the 10 x 14 side, 140 sq in at 1500psi is 105 tons of force on the press structure. The resulting block is perfectly 10 x 14 but the thickness is not perfect...A horizontal press presses the block from the 4 x 14 end. this is 56 sq in at 1500 psi = 42 tons of force on the press structure. The resulting block is a perfect 4 x 14 x 2 to 12 inches, or a 4 x 12 x 2 to 8 ", which you can determine by the simple slide of a limit switch..."

    "There are some big liabilities that go with the interlocking system that I can see. First is the dependence on cement to make them strong enough to not break the overlapping flange in the event of lateral wall movement, but the cement and shape also makes them subject to shattering under the same circumstances...Another liability to interlocking block is the difficulty in making corners, intersections, buttresses, and arches. Not to mention domes and vaults which are very important to earthquake resistant architecture as well as aesthetics...My dad, W.C Powell, designed an interlocking block press for Reickerman <?XML:NAMESPACE PREFIX = ST1 />Thailand back in 1992 and the principal was sound enough but the energy economics were not, with the increase in cost for cement and steel as well as the high cost of a much more complex machine the program was apparently scrapped."

    In-Plane Cyclic Shear Performance of Interlocking Compressed Earth Block Walls  It was determined that the shear strength of fully grouted walls is significantly higher than that of partially grouted walls and calculation of capacity based on current ACI 530-08 masonry provisions significantly overestimates the shear strength of ICEB wall panels.


  • Vote Up0Vote Down
    February 2012

    Dorkmo,  yes, it is possible to drill holes in CEBs, much like it is possible to cut them.  However, the labor and time required to do so is twice or three times as long as the process to press the blocks....effectively quadrupling the total manufacturing time. It also required additional machinery (obviously) and additional skilled labor.  These facts alone diminish the inherent value of the CEB.  This is why I so strongly advocate for a press that generates 'holed' blocks.  And if the design is being refined to potentially support other shapes of the blocks and/or interlocking blocks, then this is the time to include a template for 'holed' blocks.
  • Vote Up0Vote Down February 2012
    Hi Guys,

    Welcome to the discussion! I haven't been posting a lot lately mainly because I've been a little put off by the information we found at regarding vertical presses.

    Whilst I think we would all agree that concrete filled CEBs are an illconceived idea, that doesn't change the fact that some building codes require it (not in Australia thankfully), despite the fact you get better earthquake performance using standard bricks by changing the building design (e.g. use Domes and Vaults).

    A couple of things, first up, it's true, the interlocking bricks that I was conceiving were non-holed versions, because I didn't require it. There would be very little preventing the machine I was envisioning making that kind of brick however (as it was based on commercial machines that make that kind of brick).

    The reasons I don't want to use rebar in the construction are many:
       The Cost
       It adds arguably marginal strength given the right CEB construction techniques (domes and vaults).
       It's not fire proof (we get extremely nasty bushfires in Australia - arguably the worst in the world), the rebar would cause the walls to crack and rupture as it expands with heat (>1000 deg C or 1842F).
       And finally, I don't require it to pass any building code.

    Currently though, I have halted work on the design because I'm now not confidant that it will work, given that the height of each brick needs to be pretty exact for interlocking bricks.

    James, do you think that is is feasible for a vertical press to create bricks with the same height by varying pressure? That is the question that needs answering.

    Hey Mat I found that quote "Another liability to interlocking block is the difficulty in making corners, intersections, buttresses, and arches" pretty strange as arches, domes and vaults are exactly why I'd like to use interlocking bricks! I think that's a pretty narrow minded view don't you think? I mean I can conceive plenty of brick designs that would fullfil those requirements.

  • I figured they meant that interlocking bricks depend on their orientation for mutual strength, and corners/intersections/buttresses/arches force them to change their orientation. You'd need a few complicated pieces to properly mate one row of identical bricks to another row in a different orientation. At least, it seems to me like you would. I can't think of an interlocking design that would allow adjoining bricks to sit at arbitrary angles. The vaults seem particularly difficult. Do you have a link to an interlocking system you're thinking about?
  • >>Currently though, I have halted work on the design because I'm now not confidant that it will work, given that the height of each brick needs to be pretty exact for interlocking bricks. 

    From what I've seen, no you cannot get the type of stack-up precision (multiple variances adding up with each coarse of bricks) with a pressure or timing function. I would think the soil needs to be far too pre-conditioned to work out that way. However, as James was saying he hasn't noticed a difference between 2000psi and 3000psi bricks, what that means is that you can just use physical limit switches and as long as the soil is roughly about what you need, you should consistently get adequate bricks provided you're within the pressure range (in this case lets say 2000-3000psi). 

    You could see the difference in the consistency of the bricks between Floyd's commercial machine and the OSE CEB Press. For stacking of the OSE CEBs we just stacked them on the side that was against the walls of the chamber and not the direction of pressing. 
  • Vote Up0Vote Down
    February 2012
    @DavidIAm One thing to add would be adaptive adjustment of the downstroke position (which determines how much earth enters the chamber for compression) so that the system tracks any changes in soil characteristics during a run. In combination with your earlier suggestions this could make for more robust behavior.

    @mrsquish Re:"taking large parts of the press apart to change a brick mould" I don't think that's true; put a socket wrench on two nuts and the pressfoot comes out in one piece. If you have a specific quick-change mold system in mind, it could probably be made as a replacement pressfoot (for the bottom mold, at least) with no significant frame redesign. The drawer needs rework to support interchangeable mold plates, but again, it doesn't look to require any change to the basic design. Re: "the area that you've put the rails, there's massive draw that fills that void", it sounds like you misinterpeted my drawing. The top of the support rails (at their maximum elevation) are flush with the deck plate during ejection.


    We could modify the "lego" mold in my sketch to include a pair of thru-hole coring rods (assuming matching clearance holes in the upper mold plate). A penalty is a few inches increased length of travel required from the main cylinder/pressfoot. Lengthening the main cylinder would require some larger frame members, but not a basic design change. (I know some people think the deck and hopper are already too high, though.)


    BTW, the description I wrote at could use some review by someone who really understands what the code is doing (especially during initialization).

  • Vote Up0Vote Down
    February 2012
    On second thought, I now do NOT think the thru-hole coring rods would require extra main cylinder travel. Of course, as with the "lego" mold, the main cylinder has to retract for mold clearance before the drawer begins its ejection movement; with the thru-holes the clearance retraction is longer. But I was erroneous in thinking that the pressfoot would have to retract any lower than it does already.

    The pushrod-and-latch-pawl mechanism I sketched earlier *is* pretty complicated by OSE standards (3 independent moving parts on each side, 6 total), but I bet someone can come up with a simpler way to get the same effect if they put their minds to it.


Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Login with Facebook Sign In with Google Sign In with OpenID Sign In with Twitter

In this Discussion