Factors affecting fit and depth clearances

[Leon Rudan]

Hi. I am doing my first eCabinets project (which happens to be a large one) and am trying to understand what the fit clearance and depth clearance variables depend upon, or depend more upon.

I am going to be using a domestic (Canadian) 3/4" 120g veneer particle board core material and blind dado joinery. Unless I took my notes wrong, the company that is going to be doing my Thermwood outsourcing advises a fit clearance of 0.020", whereas I see the videos suggest it be 0.010"-0.018" and a depth clearance of 0.030", which I haven't seen a range for in the videos but have noticed is set in some at 0.015" or 0.018". I was hoping to work on my project all weekend but could not get a re-confirmation of these numbers on Friday, so I am going to call again tomorrow. By the way, they use a special glue they say they make up themselves, which no doubt there is an upcharge for. My preference is to go with a commercially available product and have chosen Titebond III.

So, back to my original question: What factors influence the choice of fit and depth clearance? I know that significant variations in the thickness of the material used require one to use a larger depth clearance as in the thinner sections of the material the bit will not cut into the material as much as it would in the thicker sections, resulting in the tenon not fully engaging entirely into the dado. But, if one's material is of very uniform thickness, as is good quality particle board core, should the depth clearance really be the same for anyone running this material on any Thermwood? For the fit clearance I would think that it should be the same regardless of whether one is using plywood core or particle board core, regardless of the variability in the thickness of the material. Lastly, does the choice of glue, which all have slightly different viscosity, affect the clearances?

I am just trying to understand this issue fully. Thanks!

[Will Williamson]

Design your self a small cabinet 12" x 12" x 12" and see for your self the result of your setting changes and flip ops

[Scott Marshburn]

As Will Stated it would be a really good idea to create a test base, upper and maybe even corner cabinets before diving into a complete library. Depending on the quantity of cabinets you have in your library it could be very time consuming and nerve racking going through and changing every thing. Trust me I have ben there.

[Leon Rudan]

Design your self a small cabinet 12" x 12" x 12" and see for your self the result of your setting changes and flip ops

I totally agree with designing a test cabinet and adjusting the clearances as required. I would right now if I could.

But, the project is being outsourced and the company is over an hour's drive away. If the machine were right here and I was operating it that would be easy. Also, the material has to be special ordered all at once as it is domestic pre-finished flat cut maple veneer particle core NAUF material (125 sheets) with a 3-week lead time. At that time I will be doing a test cabinet, which the Thermwood shop also suggested, before the entire job is run. Given I have to order the material before I can do the testing, any change to the clearances that have to be done after the testing would result in a big delay while all of the joinery is changed. If I wait to design and program all the cabinets until the test is run there still will be a big delay while I design and program everything. This delay is unacceptable to the Thermwood shop as they'll have huge lifts of material taking up space in their shop. So, I guess I'll just have to accept their clearances and hope for the best when the test unit is run and if the joinery does have to be changed I'll have to be prepared to work like crazy to change every individual cabinet if it isn't.

I guess what I could do is just design one cabinet, which I have, and ask them to run it right now on a piece of 3/4" particle board they have laying around. But that won't be the actual particle board I will be using. I don't know if that makes a difference.

In advance of that, I was just hoping someone would give me some insight here on this issue of what affects the blind dado clearances. I am trying to understand why it is not the same for everyone if the material being used is of uniform thickness.

Thanks.

[Daniel Odom]

If the material is all of a consistent thickness then .015-.02 should work fine for particle board and mdf. This is more of an issue when using plywood. The problem is that when the material does vary then you'd want to err on the side of too much slop vs too tight. I usually just use .015 for melamine fit clearance and maybe a bit deeper for depth, say .02-.03 depending on the glue/material. You should find out how thick the finished material will be if you can, save a lot of guess work and re-work. Once you know the expected thickness you shouldn't have too much to worry about. A couple of thou isn't going to make much of a difference for this material. But if the material ends up being .82 and you have everything designed with .76 then you might have to go back and rework it all.

I'm not sure about the custom glue, but it could be they have a good reason for the glue they use, maybe it has a longer setup time or is easier to clean up without ruining the finish. Standard wood glue won't stick to a slick surfaces well so it might be modified to account for that.

[Leon Rudan]

Thanks! That is exactly the kind of analysis and advice I was looking for.

I do have a direct line to the factory that is going to be fabricating the 3/4" material so I will find out from them what the thickness is designed to be before fabrication and then after fabrication. I am pretty sure they are going to tell me it is going to be exactly 3/4" thick as I've been assured by them that when they do particle board the thickness is both what they specify and consistent throughout a sheet and amongst different sheets. But, since I know it is being sent out to another company to have the veneer finished I'll ask them to check the thickness on a few sheets before it is sent there. Because of the lead time in shipping to the finishing house, the actual finishing, and then to the Thermwood shop, this will buy me a bit of extra time to change the clearances if need be. Then I will have the thickness checked again when it arrives at the Thermwood shop.

Btw, am I right in assuming the clearances that you personally use for melamine would be the same for hardrock veneer particle core, ie what the surface is veneered with makes no difference as long as they are both particle core.

Also, just so I understand this correctly, using your example: If the material ends up being .82 but was designed with .76 then the dados would be deeper than expected as the bit would be cutting first through 0.06" of material it did not expect and then the proper dado depth, resulting in a bigger depth clearance than you wanted. Alternatively, If the dado depth was, say, 0.70" thick it would be cutting the dado 0.06" less deep than required which would result in the tenon hitting the end of the dado and not engaging the part fully leaving a gap between the material with the tenon and the material with the dado.

[Daniel Odom]

If the material is a different thickness than what was programmed for then the dado depth will be off. The fit clearance wouldn't be affected as the machine is referencing the bottom of the sheet and leaving the tenon thickness specified, however the joint would no longer be perfectly flush. You might get away with it if it's just a little off, but anything over a 1/32 could cause problems. The tightness of the joint is more likely to be influenced by waste-board thickness and tool length accuracy, bit flex and material flatness. The fit clearance can compensate for the variance in this regard. So for .01 to work everything would have to be set up dead on and you're machining perfectly flat material; in reality somethings always going to be a little off and you need a little room for glue too. It's also easier to assemble somewhat loose joints as opposed to perfectly tight.

You can update material thickness before sending to machine, so you don't have to go back and change every cabinet. Should work without any problems as long as the difference is only slight. If the difference is over a 1/32 you might want to update the material thickness and write a twd for the box parts but create another seperate twd without updated thickness for the door and drawer fronts if you're doing frame-less construction and verify that the door sizes to be cut are correct on the machine.

Is the material going to be finished on both sides? Might want to find out as I've run some cabs from pre-finished ply on one side and it was all I could do (or couldn't do) to get the material to vacuum down flat. Maybe it's not as much of an issue with pb core.. hopefully not anyhow. Or is this just a melamine finish on one side?

[Leon Rudan]

If the material is a different thickness than what was programmed for then the dado depth will be off. The fit clearance wouldn't be affected as the machine is referencing the bottom of the sheet and leaving the tenon thickness specified, however the joint would no longer be perfectly flush. You might get away with it if it's just a little off, but anything over a 1/32 could cause problems. The tightness of the joint is more likely to be influenced by waste-board thickness and tool length accuracy, bit flex and material flatness. The fit clearance can compensate for the variance in this regard. So for .01 to work everything would have to be set up dead on and you're machining perfectly flat material; in reality somethings always going to be a little off and you need a little room for glue too. It's also easier to assemble somewhat loose joints as opposed to perfectly tight.

I haven't heard the term waste-board but in googling am I right that it seems to be a board you put on the CNC that the actual board you are cutting goes on top of? Even reading the threads on here about waste board I am having a hard time figuring out what it is. Nothing comes up when I enter it into the eCabinets search field in Help.

I didn't realize the machine references the bottom of the sheet being cut when it cuts the tenon but that would be the only way for the tenon to always be the correct thickness regardless of variation in the sheet thickness. For the dado then does it also reference from the bottom of the sheet back up to the height of what the thickness of the sheet is programmed to be and then cuts downward from there? ie therefore resulting in a dado that is not as deep as it is supposed to be if the sheet is thinner than the machine thinks it is?

You can update material thickness before sending to machine, so you don't have to go back and change every cabinet. Should work without any problems as long as the difference is only slight. If the difference is over a 1/32 you might want to update the material thickness and write a twd for the box parts but create another seperate twd without updated thickness for the door and drawer fronts if you're doing frame-less construction and verify that the door sizes to be cut are correct on the machine.

Is the material going to be finished on both sides? Might want to find out as I've run some cabs from pre-finished ply on one side and it was all I could do (or couldn't do) to get the material to vacuum down flat. Maybe it's not as much of an issue with pb core.. hopefully not anyhow. Or is this just a melamine finish on one side?

The cabinet material is pre-finished only one one side. It is domestic, ie made-in-Canada (Quebec actually) 3/4" thick pre-finished flat-cut maple veneer with 120g particle board core with both the finish and the core being VOC-free and NAUF. The other side will be A3 or A4 quality which means a lesser quality veneer will be used on the other side as it will not be seen. The doors will either be MDF or pb core, slab design, veneered on both sides and supplied by the company that is doing the custom veneer pressing with sequence-matched American Walnut so that the grain is continguous and consistent across the door and drawer panels. I am not sure yet whether we are going to cut all of the doors and drawer faces on the Thermwood because I think the kerf size on the CNC drill bit might be too wide and therefore ruin the look. So, I may have the company that is custom pressing the veneer do the cutting which the don't do on a CNC but I imagine they do on a table saw.

Not to get into a discussion about ply vs pb as everyone has their personal preferences, but I am using the particle board core, even though it is heavier than ply (and I am using full tops) because of advice I have gotten from a cabinet maker whom I really respect which is:

- It is generally less $$ per sheet for the same face veneers (definitely the case with what I am using);
- is very uniform in thickness;
- actually has a better face impact resistance, due to the face of cabinet grade particle board being case hardened;
- machines cleaner and there is less chance for grooves, rabbets and dadoes to have issues with chips, etc; and
- accepts screws and nails better, usually without splitting and less chance of face veneer to chip off around cuts, end, etc

I am curious. What was the reason you were you not able to get the pre-finished ply to vacuum flat. Was it because it was slightly bowed?

[Daniel Odom]

Oh, I'm not knocking the material, I was just thinking about the material being bowed on the side you're trying to machine, b/c it's a p.i.t.a. when that happens. I've never used veneered pb only veneered ply and mdf core, all the pb I run is either melamine or laminated pb. Yes if you only finish one side of a sheet of ply it can warp pretty bad. You also might be creating a lot of flip ops by using tenon joinery on one sided material which could increase machine time and handling so that's something to consider anyhow.

You are correct about the wasteboard and z reference.

[Leon Rudan]

Thanks. The veneered pb core I am using has a lesser quality (A3 or A4) veneer on the flip side. I thought all veneered pb core or plywood core did. Plywood core has a greater tendency not to be of uniform thickness (esp the 'China Ply') and to warp and twist while stored,.

In order for me to understand how the machine works better I would really appreciate it if you or maybe someone else) could answer the 3 questions I had posed before your reply, which were.

1. What is the 'waste board' and it's function?

2. For the dado then does it also reference from the bottom of the sheet back up to the height of what the thickness of the sheet is programmed to be and then cuts downward from there? ie therefore resulting in a dado that is not as deep as it is supposed to be if the sheet is thinner than the machine thinks it is?

3. What was the reason you were you not able to get the pre-finished ply to vacuum flat. Was it because it was slightly bowed?

Thanks!

[Daniel Odom]

1.) wasteboard is just a spoilboard on top of a spoilboard basically. The top of the waste-board is zero in most cases but is offset by the material thickness when running a sheet so that the top of the sheet becomes zero and all z values are negative. What it all means though is that the waste-board needs to be flat and ridgid uniformly to be accurate and even then it can still be off by a .003 +- and can fluctuate throughout the day depending on humidity. Some people use a new thin waste-board sheet for every sheet run and some people just mill down the waste-board every couple of sheets run to counter this.

2)yes


3) If a material has a "frown" type bow with good side up it will vacuum flat. If it has a "smile" type bow with the good side up you have to fasten it down or some other method to get it to suck flat. Some of the cheaper ply's have a pringle bow which you can just forget about running it unless you have a roller hold down system or a box of double sided tape, a 5-in-1 and some time to kill. Finishing one side will make it bow in one direction but a factory supplier may have ways to keep this from happening and it might not be as much of an issue with pb core.

[Leon Rudan]

Thanks. I am understanding this better all the time. But, of course, I have a few more questions I hope that can be answered:

1. I'll have to take a look at the CNC cutting my test cabinet when I am at the Thermwood Shop next month as I thought the material being cut sat right on the flat bed and was kept in place by downward suction. This is a rudimentary (and maybe stupid question) but If there is a waste board underneath the board being cut what keeps the board being cut from moving? I can't really tell from watching the video's posted on Thermwood's Youtube channel.

2. I reconfirmed that my Thermwood shop uses 0.020 for the depth clearance and 0.030 for the fit clearance. I thought the fit clearance was always less than the depth clearance, or at least it is in Thermwood's training videos and whenever I've heard anyone talk about them on this forum. Or, is this not always the case? I asked for an some explanation of why but the answer was that they just always used those numbers.

3. What governs the tenon thickness as a percentage of the material thickness? The default in eCabinets when you load a standard cabinet is 55% but my shop uses 50%. Either seems to make sense and I am using 3/4" thick material. But, what is really the criterion for deciding? Again, when I asked they just said it's what they've always used.

4. The shop advised me to use a 1.25" inset for the front of the blind dado on the tops and bottoms and 1" for the back and was told that that makes it easier to identify what goes where during assembly. But, since my cabinets are all maple veneered on only one side (with an A3/A4 veneer on the other), each part will be labelled and all the parts for the each cabinet will be flat together for shipping to me so I don't have to search through a pile of parts for the one I want during assembly, it seems to me that using different insets isn't really necessary. But anyway, I did ask about using assembly marks as I was curious and they said they never use them and that the different insets was an easier way of doing it. In what situations do people on here use assembly marks? Are they just for really complex cabinets?

Thanks again for everyone comments.

[Daniel Odom]

Without getting into the technical explanation of atmospheric pressure, blah blah blah, all you need to know is the vacuum works through both spoilboards no problem but in order for the sheet to hold down effectively the surface of the sheet needs to make contact with the spoilboard before the vacuum seal kicks in and holds it all down.

.03 fit clearance means plenty of slop in the joints. Should work fine though I usually go a bit tighter than that. Of course, I'm able to test fit as I'm running and am able to make adjustments as needed. Once all is glued, clamped and fastened those numbers should work fine. The percentage may depend on the diameter of tool used to cut the dado. For example if you wanted to use a 3/8 bit to cut the dados then 50% or more of .75 thickness material would do that. It's not really related to the depth clearance.

Couldn't really say about the insets, if the part is symmetrical I'd keep it the same. Is there a dado or something that has to be flip op'd? Maybe that's so the operator knows which is end is the front.

Anyhow, sounds like the shop you're dealing with knows what they're doing so I wouldn't worry about it too much. Just make sure everything gets labeled correctly and checked off as they go. I only use assembly marks for complex cabs so I know how it goes together later on. I used to try and use them on all the cabinets but the parts would still end up assembled incorrectly, even with labels and illustrated instructions. Now I just pile up the parts according to cabinet and demonstrate an assembly of one.

[Leon Rudan]

Regarding the different insets on the opposite ends of the blind dados, no there is no flip operation required and the parts are symmetrical as they are just full tops, bottoms and backs of standard cabinets. Also, they are even flat packing all of each set of cabinet parts separately. I am loathe to ask why they suggested different insets given the previous "we always do it that way" reasoning they have given me. I always like to know how decisions are made and why so I can better understand the process I am a part of.

The fit clearance they use would result in plenty of slop. Now I am wondering if they simply got the depth and fit clearance reversed from the very beginning and since then always 'did it that way' (their stock answer for every other explanation I requested) as I've never seen an instance where the fit clearance was greater than the depth clearance and 50% more at that, ie .030 vs 0.20.

[Dennis Englert]

Re: Different insets. A partition or shelf with a blind dado could be installed incorrectly. For example, if a partition was flipped then the opening to one side is now wider, while the width of the other side is reduced. The different insets should make it nearly impossible to flip the part. Assembly marks would solve that problem.

Dennis