Door Sanding Quality Issues

Here's advice on setting quality standards for cabinet door final sanding, and meeting the goals in a competitive market. April 2, 2013

We are a new cabinet door company and struggling to stay afloat. When we started our business plan the plan looked great. There was not another door company within 150 miles and a ton of cabinet shops and custom woodworking shops in our area. We have been struggling with our workflow and it seems to be taking forever to get a single small job of about 25 doors out in a day. We have found some of our bottlenecks and fixed what seems to be all but one. We were running all our stile and rails on a unique 250 door machine but that feels extremely slow.

We have now been running all our stiles on our moulder and that has helped production tenfold. We are now dealing with a huge problem in our sanding. We currently have 37x60 wide belt sander but are having to sand out all the cross grain scratches and itís eating up a whole day. We have had to turn down some shops that have such a high volume of work that we just couldnít keep up because of our finish sanding problem.

What are we doing wrong? Is this normal to have to spend so much time hand sanding doors? We are going to try and use a mop sander to sand the cross grain scratches. Has anyone tried this method before? With all the equipment we have we should be able to produce far more than we are at this time. We have a ton of money in this but some days you want to say the heck with it and working 16 hour days and not making a dime for your family doesnít make you bright eyed and bushy tailed in the morning.

Forum Responses
(Cabinetmaking Forum)
From contributor X:
What grit are you sanding the doors to from the widebelt? Are you using pneumatic DA's after that? If so, what grit are you using for that? Do you think the widebelt is the problem? Maybe it's not doing a good enough job and creating too much work for the hand sanding.

From contributor Z:
Depending on the market that you normally service, it might be possible to allow cross grain scratches to stay in your product. You might consider discounting in such a case for some clients that do not require a more perfect product. Saying that, I've found that when dealing with cross grain scratching from a wide belt sander that it is advisable to final sand on the widebelt with a fine belt such as 180 grit before final sanding with a DA. Start your DA sanding with the same grit or a grade slightly courser and then proceed to final finish sanding. Hand sanding should not be necessary except for breaking the edges with a flexible sanding sponge. The key for me is to not use excessive force when using the DA and to not jump up in grit too fast. You might also consider letting cross grain scratches stay in the product on the insides of doors.

From the original questioner:

I have been starting out at 120, then using 180, but I felt like I needed to use 150 after the 120 and finishing at 180 on the wide belt. I am using air operated DA sanders using 180. Before the wide belt was giving us really bad track marks that looked like a dotted line, so trying to sand that out was a very time consuming, but after doing some research we lowered the air pressure and that seems to have solved that problem.

Shops around here expect their doors to arrive paint room ready and I agree. Is this a SOP for sanding on the wide sand and hand sanding? Is my WB sander performing bad? Or do I just not know what I'm doing? Shops that we have done work for say our quality is great but we are paranoid that we may send something out that hasnít been sanded properly, and our work flow is extremely slow in our sanding area. Is it true that you should be able to take the doors from the WB sander to the paint room? I donít see how this would ever be possible.

From contributor H:
I know one of the most expensive door companies around here sends every door out with cross grain scratches and sharp edges and bowed stiles. Then add on labor to sand and they become super expensive doors. Quality takes time.

From contributor Z:
Does your wide belt sander have a platen between the drums? If it does, you might check to see if it is set correctly. I would also check your drums for imbedded debris or missing rubber. After checking these things I would suggest making sure that the belts you are using are in good shape. I always made sure that they were free of glue residue and filler that might have not dried sufficiently before sanding. Check your belt after a sanding procedure for this as it's better to get it cleaned up before drying occurs. I would also check the oscillating function of your drum.

From the original questioner:
To contributor Z: The WB does have a platten and I have it set to the lowest position because I believe I am getting the best results. I am using a new belt and getting the same results. The oscillation seems to be working. I can eye the belt moving from side to side - is this correct? The rubber contact drum seems to have a bad spot that is about four inches wide but I am getting the same finish across the whole belt. When we send the doors to the WB they are 13/16 and we sand to finish 3/4. I believe that if we can get this problem under control we would be in good shape.

Can anyone share their process from rough stock to a finished door? The following equipment we have: Wienig 5 head moulder, 24in upcut saw with Tiger Stop, eight foot JLT panel clamp, JLT door clamp, unique 250mc door machine, Extrema SLR, 20in planer, 15in planer, Midwest 37x60 wide belt sander, Accu-System MMTJ miter door machine (which we are installing now) and a panel saw. We are using insert tooling as well. With this equipment we should be making money but we arenít.

From the original questioner:
I have faith in our ability to make a great product. We have been in production for only two months and we haven't received any complaints at this point. Maybe we are being too hard on ourselves too early in the game. It was mentioned to remove 13/16 to 3/4 of material. Is this a lot to remove? What thickness should I be taking to the WB?

From contributor Z:
Here's a pretty good Knowledge Base article about adjusting the platen on a wide belt sander that might be helpful to you.

Widebelt Sander Adjustment

From contributor D:
So is the equipment new or is it used and paid off?

From the original questioner:
Most of the equipment is used but is in very good condition. It is not paid for yet. It is hard to come up with that amount of cash to purchase all the equipment. We have about $6k just in tooling and thatís really not enough. On my WB control panel I donít think I can send it through at .781. I think itís what you are doing? I have to put in .75 for 3/4 and so on. I donít think I can use a three digit number. Would this make a difference? How many doors (raised panel) are you putting out in a day from rough lumber to finish product?

From contributor T:
Is there any possibility of solving the cross-grain sanding problem at the source? Could you widebelt your sticks before machining and then pay better attention to cope/stick alignment during glue-up? If you did this could you produce an adequate final sand with just a DA sander? How many of your doors are raised panel? Could you offer a rapid turnaround on shaker style recessed panel? If you didn't have so many set up costs could you distinguish yourself with a smaller product offering but outstanding turn-around time?

From contributor R:
When I was buying doors, turnaround was paramount to cashflow, and everything fed off that schedule. I read a post on here a while back that said "completed beats perfect every time" - there is some truth to that. I can't tell you that I would sell a blatantly defective product, but pretty darn good gets you paid and everybody else happy 99% of the time.

From contributor O:
I can't say I have the solution to your valid questions. Your lack of confidence about where your product needs to be leads you to perhaps overwork the product and deliver more than is needed. You need to resolve this insecurity. The solution is to gain confidence by talking to your customers and have three doors, all processed differently to show a range of cross grain scratches. Ask the buyers what they fairly expect to see for what they are paying. This will help you know what you need to deliver, to produce. Buy a few of your competition's doors and look at them closely. This will give you a perspective.

Then tune in your sander. With that equipment, you should be producing everything much closer to final thickness, and then fine and super fine sanding to thickness, then a quick R/O to finish. Why not final sand to 250, which will not show cross grain scratches, then R/O to 150?

From contributor F:
I'm a one man shop so Iím not up to the level of output youíre dealing with. However I do see one thing I would do differently. You say youíre going from 180 grit on the widebelt to 180 grit on the air sander. That's going to take too long! You want to drop down to 150 on the DA sanders to quickly remove those cross grain scratches. In reality you probably don't need to go back to 180grit either. Using the 180 on the widebelt is good to get the scratch pattern as fine as possible. Then a quick, (though thorough), sanding with the DA at 150 should get you a pretty good finish. I don't think I've ever sanded a cabinet door to 180, though I know there are some guys out there sanding to 220 so you have to decide what's good enough.

From contributor X:
Personally if I was in your situation I would be using 120 grit with a DA (after they come off the widebelt).That will save a lot of time sanding off crossgrain. Then I would let the customers know it's up to them to do the final sanding prior to their finishing since everyone has their own preference for sanding grits prior to staining. The customer should be responsible to final sanding anyway since they can get marred or scratched from handling them from your shop to the shops. I think 120 grit is plenty good to make them presentable, knowing they will be doing the final prep work anyway.

From contributor W:
To the original questioner: How many doors are you currently turning out in a day? What were your expectations as far as production capacity? Making strictly doors is a volume based business. You can't make any money by turning out a few doors a day, which I am sure you have realized by now. Going by the equipment you mentioned, I think maybe you are only halfway there with what you really need to turn out any kind of volume.

As far as your sanding problems go: I take it you have a single head widebelt sander. Nothing wrong with that, but it simply means that you need to run the doors multiple times through the machine with different belt grits. You need to start off with an 80 grit to knock down the doors to rough thickness than maybe 120 grit, then 180 grit from there to finish sand. You shouldn't be taking more than 20 thousands on the finish pass with the 180 grit as more will load the belt and limit the life of it. Also hitting any staple pins will leave a mark on the belt which will show up as a raised surface line on the door. Once that happens to the belt it is there for the rest of the belt life. As business increases you should start looking for a two or three head widebelt sander. One pass, front and back through a multiple head sander will reduce your labor considerably.

As far as the finish quality: You really need to look for a random orbital wide belt sander. Timesavers makes a few models. There maybe a few other brands as well, but Timesavers is what I work with. I have seen them with a standard single and two head widebelt then either single or double orbital sanding heads. They also come with just the orbital sanding heads as a standalone machine. They are not low cost machines however. The machine basically does what you are doing with the pneumatic hand sanders except much better and faster then you could ever do by hand. Most of the guys running the orbital sanders are running 220 grit pads. Any higher grits will polish the wood and make getting a consistent stain finish hard.

From the original questioner:
Contributor W: I was thinking that we should be able to turn out 50-100 doors a day - is this reachable? I delivered a set of oak doors to a shop today and the owner asked why did I finish the doors so smooth (180 grit). My response was I wanted to make sure it was perfectly smooth without any cross grain scratches. He said that I may have finished it too smooth that he would have to stain them and see. I had one shop call and asked if we could finish to 320 grit so I am left scratching my head at this point.

The WB is a single head sander. I believe I am going to try some of the tips that I have received on here and maybe I can find a happy medium. I had a guy start with me today and he seems like he is going to be a big help with speeding the process up. I feel that if I can get this sanding issue solved the future looks bright. We are going the try a mop type sander to see if it will get the cross grain scratches out any faster.

From contributor F:
You need to offer a basic door that's going to be sanded to ďXĒ grit. Then if Customer A wants a door sanded to 220 grit it's an upcharge of ďXĒ dollars. Customer B wants it sanded to 320 grit. It's an upcharge of money. This way you have a standard door which is acceptable to most and can still, (assuming you want to), provide additional sanding options - for a price. I personally can't think of any reason to ever have a cabinet door sanded to 320 grit. I personally wouldn't go over 150 grit unless someone asked for it but you'll have to decide what grit works best for your clients. If you look in the box store cabinets you always find cross grain scratches on the inside of the doors!

From contributor I:
The mop sander may work with some dense wood, but on oak or ash it will probably sand the early wood deeper than the late wood. Giving you texture on the surface. Stick with the random orbit sanders, and use the coarser grit to start as mentioned.

From contributor W:
I don't see why you couldn't turn out 50-100 doors a day. Of course it is hard to know if that is possible without knowing how you process the work. Making flat panel doors is a different process than raised panel or miter doors. There is a lot more work that goes into a raised panel door to make the panels than just cutting some plywood or MDF for the flat panel ones. I wouldn't sand to more than 180 with the widebelt and no finer than 220 with a random orbital. You can of course do what others have suggested and backsand with a lower grit after widebelting.

You are still making scratches no matter what you do - trading off the crossgrain scratches for swirl marks with the orbital sander. All of which will show up to some extent with a dark stain. We haven't had much luck with brush type sanders as they are more for de-nibbing. They have a way of rounding edges off. Sanding quality to someone that is going to paint them would be different then to someone who is going to stain them. What about profile sanding of the end grain of the raised panel or outside edges?

From contributor Y:
We are a production cabinet shop, 4,000 cabinets plus a week. We have three sanding lines for our doors using a DMC widebelt followed by a DMC orbital sander. It removes 99% of all cross grain scratching.

Click here for higher quality, full size image

Click here for higher quality, full size image

Click here for higher quality, full size image

From contributor M:
As a cabinet shop that outsources doors I prefer a final sand at 150 grit. Sanding too far will affect how well my stain is absorbed. I rarely sand any finer than 150 until after sealer coats.

From the original questioner:
Following up with much delight I think with all the help we have found a new but old way of doing things. We just finished an order of 48 doors and 26 drawer fronts. We sanded from 100 grit to flush all joints (two passes front and back) then one pass with 150 front- back and the 180 front-back (one pass) and with the WB sander and then orbital sand with 150 to remove the cross grain scratches. The mop sander didnít make any improvement except to remove the softer grain which made the wood feel textured, but we will use it to sand mitered profiles and moulding.

Although this process has made the process from start to finish faster I still wish there was a faster process. I forgot to mention that we do have two profile sanders for sanding the end grain on our raised panels which makes a nice finished panel. After taking the advice on not finish sanding over 150 as a standard is a great idea and we have now implemented this as a standard. If requested we will offer other finish sanding grits at a slight up-charge. We are still cutting our teeth at this moment but I know we will be ok.

From contributor M:
Something to consider in your WB sanding: Consider the scratch depth of the 100 grit belt. I did not do the math on it but I donít believe one pass with 150 will take out the scratch depth of 100 grit, and one pass of 180 will not take out the scratch depth of 150. You then are trying to take out cross grain scratches of a 150 grit with your random orbit sander. With our WB we have a twin head and we go one pass 100-150 each side or until smooth and then go 180-220 once and the 220-220 once. We then random orbit down to 180 or 150 if someone requests and we are only needing to take out 220 scratch depth marks instead of 100 or 150.

From contributor A:
Yes 120 will work for a first pass if your parts are flush. Try 120, then 180, and then orbit with 150. If you have trouble taking the 120 scratches out slow down the feed a little.

From contributor G:
Get a hold of a Festool RO (150 or 125) with a variable suction vacuum extractor on it. First run rotary (actually sort of rotary) with 120 grit, then switch to random orbit. Then slap on 150 grit and then do rotary and then random orbit. Blow of the dust between switches of modes and papers. Adjust the vacuum to just pull it down to the rails to get the job done at each step. Too little vacuum will cut slower, too much will cut aggressively. This will go fast once you get the hang of it and the job can be delegated to almost anyone who will look at what they are doing. I agree with all of the above comments on sanding too fine and the stain won't soak in.

From contributor E:
Too fine a grit as mentioned will prevent stain from coloring the wood properly, but another issue is adhesion of top coats. Some manufactures suggest not sanding past 150 grit to avoid adhesion problems. I would not sand past 150 on finish sanding and let your customers know why.