when to use the wide belt platen

[David Zaret]

i have a wide belt with a steel drum and a platen. my thinking has always been to "use the platen for final sanding." recently, a friend told me "no, use the platen for all sanding, except the final pass, which should be the drum only."

what's the science and thinking behind the correct approach? ... and what's correct?

thanks,

-- dz

[Erik Loza]

I am not an expert on WB’s but was instructed to always use the platen, no matter the task. If you think about it, a WB with no platen is basically a glorified drum sander. Most of the shops I worked with just switched back and forth between felt and graphite platens, depending on the job. I realize this isn’t a clear explanation to your question but using the platen is pretty much SOP in the industry.

Erik

[Warren Lake]

curious what the logic is behind drum last pass. A drum is one small contact point compared to a platten. Wide belt i have is a drum. Other than for thicknessing my stroke sander blows it away.

[Kevin Jenness]

The platen is for finishing passes. It lengthens and softens the scratch pattern but runs hotter. Drum only is more aggressive for rapid stock removal.

[David Zaret]

The platen is for finishing passes. It lengthens and softens the scratch pattern but runs hotter. Drum only is more aggressive for rapid stock removal.

thanks - this is consistent with my thinking as well. guess my friend was wrong.

-- dz

[Kevin Jenness]

The question is easily tested. Take a moderate cut with the drum, lower the platen as you go and watch the ammeter climb. Now take off an equal, light amount of material sufficient to remove the previous grit scratch pattern using the platen on one sample and the drum only on another. See which one sands out more easily with a random orbit finish sander.

A wide belt without a platen is not a glorified drum sander. The belt runs much cooler and loads up less because of the belt length and oscillation, plus grit changes are far faster and easier.

[Warren Lake]

not really a wide belt with a drum is still a wide belt, it has a belt. A drum has the paper wrapped around the drum and is mickey mouse. The natribom also oscilates. Old quality machinery Italian stuff. Takes seconds to change the belt.

They are referred to as calibrating machines more than sanders which is logical as they do that better than sanding.

[Mark e Kessler]

That goes against everything I have learned from widebelt sandpaper manufacturers (sia, vsm, uneeda, klingspor) and widebelt techs, (scm, viet) and research on my part. Essentially platen is for finishing 150 and above, drum is for removal 120 and below. The drum leave short scratches which can effect finishing especially when staining. The platen extends the scratches minimizing andy finishing issue from the widebelt.

The platen shouldn’t take of more than 0.002 and the platen needs the proper density of felt and that is around 65, (SCM spec is 68) if you go harder you will potentially transfer belt slice marks (aka chatter). I use 26S1 felt which is 55-65, when you glue it to the platen (the metal part) and before you put the graphite on you need to calibrate it through the sander with 100 or 120 or 80g and an angle of no more than 10-12deg (so it doesn’t flip over) this insures it is parallel. There is also a correct thickness but it is machine dependent so check that. The graphite i use is PEC 500, you can use any but the 500 does not transfer graphite to the belt.

As far as belts go, there are different joints and typ have to be requested or you will get the standard “t” joint, that is fine on the coarser belts (120 and under but not great on 150 up especially if you have a steel head like i do. You want cloth on 120 and below and paper on 150 up, on the paper you want to request a “zero tolerance” joint to minimize belt splice transfer (chatter). My first choice for belts would be SIA but it was difficult to buy from them and lead time was out of control I am using Klingspor and the customer service is great and their paper is good. Sandpaper is like beer, life is to short to sand with crappy paper so don’t cheap out…

no most of this is purely “by the book” kinda info but i like to have a baseline as a starting point and adjust based on experience, machine ect…

Also there is a min and max to sand and to remove previous scratch otherwise you are wasting time and sandpaper…but thats a whole other can of worms…

[Warren Lake]

the wide belt i have only has a drum. Thats why Ive seen it referred to as a calibrating machine. It lines up to what you said about the drum only I was pointing out that it is not paper wrap on the drum, that belt changes are fast and it oscillates. Steps up from Drum Sanders. I bought it specifically to thickness on one job when material came in over spec. I only tried it for sanding once or twice.

[Mark e Kessler]

the wide belt i have only has a drum. Thats why Ive seen it referred to as a calibrating machine. It lines up to what you said about the drum only I was pointing out that it is not paper wrap on the drum, that belt changes are fast and it oscillates. Steps up from Drum Sanders. I bought it specifically to thickness on one job when material came in over spec. I only tried it for sanding once or twice.

A steel head like the one i have (sandya win) is marketed/sold as a calibration sander were as rubber heads are sold as finish sanders but the rubber on sandya’s of this era (2000’s) are 85 durometer, nearly as hard as the steel heads. Ideally you want a max 75duro for a finish sander. Going further , a 2 head - 1st steel, 2nd rubber(75duro max) with platen (a combi head) or variations of that work well

[Joe Calhoon]

Yes, drum for calibrating and platen for finishing in higher grits. But there are many exceptions to this rule. For example when sanding dovetail drawer parts before dovetail joinery I don’t use the platen because it can have a slight rounding effect at the edges that can cause joints not to close. You will also find a lot of controversy about drum vs platen among pro shops. Some claim they get a better scratch pattern without the platen. This is only true if you have a rubber roller. With a steel roller probably better to use the platen. Multi head machines usually have steel first roller then rubber second and third heads. Also with different hardness.
when they first set up my Kundig they said paper belts were the best way to go. In practice this did not work well for us. The paper is very fragile and we were going through belts before they were worn out. Also if you have a single head machine and changing belts the paper will not hold up. I Started with SIA paper but use Klingspore cloth with paper on the last head because it never gets changed. In solid wood the difference between paper and cloth is very slight. Solid wood sanding is very forgiving but if your machine is veneer capable the paper selection and setup can get very technical.
Best to experiment with your machine to see what works best for you.

[Warren Lake]

the natribom has a rubber head and is considered a calibrating sander. Seen it written a number of times.

German company had me on cloth belts on the stroke for many years. 3M switched me to paper on the stroke they were fine. There is no constant pressure on a stroke like on a wide belt. The belts are also longer at 310 inches on that machine so for both reasons run cooler. I never tried paper on the wide belt, it didnt make sense that it would be strong enough. 3M cloth on the wide belt. Went from cloth to paper on the edge sander and that worked fine as well.

[Warren Lake]

on a stroke you hang your clamped up drawer off the table and sand it after its glued up and dry. Works very well.

[Joe Calhoon]

I ran a stroke for years Warren. Great for the outsides after assembly as you describe but not good for the inside because they are not calibration machines. They take a lot of real estate and would have kept it if more space.

[Warren Lake]

always sanded the insides first no issues. You control where you put the pressure. If the inside was rounded off on the outside edges on a stroke then its operator error. Yes 12 feet of space but can do stuff wide belts cant.

[David Zaret]

Yes, drum for calibrating and platen for finishing in higher grits. But there are many exceptions to this rule. For example when sanding dovetail drawer parts before dovetail joinery I don’t use the platen because it can have a slight rounding effect at the edges that can cause joints not to close. You will also find a lot of controversy about drum vs platen among pro shops. Some claim they get a better scratch pattern without the platen. This is only true if you have a rubber roller. With a steel roller probably better to use the platen. Multi head machines usually have steel first roller then rubber second and third heads. Also with different hardness.
when they first set up my Kundig they said paper belts were the best way to go. In practice this did not work well for us. The paper is very fragile and we were going through belts before they were worn out. Also if you have a single head machine and changing belts the paper will not hold up. I Started with SIA paper but use Klingspore cloth with paper on the last head because it never gets changed. In solid wood the difference between paper and cloth is very slight. Solid wood sanding is very forgiving but if your machine is veneer capable the paper selection and setup can get very technical.
Best to experiment with your machine to see what works best for you.

joe - thanks for this. maybe we can talk more about this in person in november when i see you out in CO. i talked to the Kundig guys about an upgraded wide belt.... it's something i'm half-considering.

[Mark e Kessler]

Yes, drum for calibrating and platen for finishing in higher grits. But there are many exceptions to this rule. For example when sanding dovetail drawer parts before dovetail joinery I don’t use the platen because it can have a slight rounding effect at the edges that can cause joints not to close. You will also find a lot of controversy about drum vs platen among pro shops. Some claim they get a better scratch pattern without the platen. This is only true if you have a rubber roller. With a steel roller probably better to use the platen. Multi head machines usually have steel first roller then rubber second and third heads. Also with different hardness.
when they first set up my Kundig they said paper belts were the best way to go. In practice this did not work well for us. The paper is very fragile and we were going through belts before they were worn out. Also if you have a single head machine and changing belts the paper will not hold up. I Started with SIA paper but use Klingspore cloth with paper on the last head because it never gets changed. In solid wood the difference between paper and cloth is very slight. Solid wood sanding is very forgiving but if your machine is veneer capable the paper selection and setup can get very technical.
Best to experiment with your machine to see what works best for you.

Joe i was told by VSM not to use paper with the steel head but Klingspor and Sia said ok at lower pressure, i think i set mine at 45-55 for paper, 55-65 for cloth and 65+ for polyester. The SCM tech told me all new widebelts they sell ship with paper (not that that means anything). I haven’t had an issue with the paper but my belts are only 25 x 60 and it’s only me so i am careful, if i was full time with employees and a one head i would be cloth all the way and might even switch when the paper belts are done, like you said the difference is slight

[Bobby Robbinett]

joe - thanks for this. maybe we can talk more about this in person in november when i see you out in CO. i talked to the Kundig guys about an upgraded wide belt.... it's something i'm half-considering.

Those Kundig wide belts are very nice but also very expensive. I believe a 2 or 3 head unit is over $100k. (I could be wrong). We run a Cantek and while it’s no where near as nice as a Kundig it has been a great machine for me. A huge step up from a Grizzly. If I had $150k+ to spend on one I would get one of those 3 head Biessee’s with the orbital robot arm. Or a machine with an orbital station to pretty much eliminate any sanding by hand. Another significantly cheaper option that I have considered for sanding cabinet doors is to add a Brush sander. I can get a Giantway 5+ station multi brush sander for around $18k delivered to my facility. These things are a wonder for sanding profiles, door edges, and sealer sanding at all levels. Speaking of which Giantway also makes a specific lacquer sanding wide belt and a veneer sanding wide belt with a segmented platten. Both are very affordable for what they are. I could buy a 5+ station multi brush machine and a lacquer sanding machine to add to my Cantek and still be at less than 1/2 the cost of a Kundig. Just food for thought.

[johnny means]

I've done a lot of experimenting here. Thousands of peices. I landed on a system where 40 and 80 grit are done with the roller any thing beyond is done with the platen down. My experience is that anythihng beyond 80 just cannot remove enough material to remove the chatter marks (scallops) from the previous grit. My sanding regimen aims for no more than one pass at every grit after flattening is acheived. Also, my last grit is always a paper belt while the others are cloth.

[David Zaret]

Those Kundig wide belts are very nice but also very expensive. I believe a 2 or 3 head unit is over $100k. (I could be wrong). We run a Cantek and while it’s no where near as nice as a Kundig it has been a great machine for me. A huge step up from a Grizzly. If I had $150k+ to spend on one I would get one of those 3 head Biessee’s with the orbital robot arm. Or a machine with an orbital station to pretty much eliminate any sanding by hand. Another significantly cheaper option that I have considered for sanding cabinet doors is to add a Brush sander. I can get a Giantway 5+ station multi brush sander for around $18k delivered to my facility. These things are a wonder for sanding profiles, door edges, and sealer sanding at all levels. Speaking of which Giantway also makes a specific lacquer sanding wide belt and a veneer sanding wide belt with a segmented platten. Both are very affordable for what they are. I could buy a 5+ station multi brush machine and a lacquer sanding machine to add to my Cantek and still be at less than 1/2 the cost of a Kundig. Just food for thought.

bobby - my issue is power. the reason i was considering a Kundig is because they can customize the configuration, and the details, such that i can operate it on my digital phase converter. i do not have 3ph from the street (power provider wanted $85k), and that's a huge limitation. so the really massive machines are off the table for me, but Kundig could create a machine that i could run, even multi-head. yeah, they are very expensive, but perhaps it's an upgrade path.

i'm running a 37" single head Stiles machine now, and it's fine. it has digital table control, seems to keep true, and just runs. it was... $18k or so 5 years ago, so very inexpensive. nothing exciting, but it sands, i suppose. i have a Kundig Uniq, and love it, and covet a wide belt... but it's unclear if that's a worthwhile place to put $100k.

-- dz

[Warren Lake]

solid off the planer can be sanded with 150 or 180, or if im doing a lot could be 120 then 180, or 120 and then 220, whatever depending on, wood, quantity and if new or used belts.

Why is it you need to go to 50, and 80 and whatever else grits?

[David Zaret]

i do a lot of veneering, and often have to size the substrate to a particular thickness. i do that on the wide belt with a coarse grit.




solid off the planer can be sanded with 150 or 180, or if im doing a lot could be 120 then 180, or 120 and then 220, whatever depending on, wood, quantity and if new or used belts.

Why is it you need to go to 50, and 80 and whatever else grits?

[Warren Lake]

thanks David, that makes sense,

I dont need to do that with the stroke as its not set to a given thickness so all I do is sand the veneer once its laid up or if i bought cut to size same thing sanding that. The only reason id go coarser on the stroke if i did a 40 plus inch table top and need to level joints board to board. 100 is more than fine for that or if only one top even 120 would be fine. Its different each time depending.

[David Zaret]

johnny - how much do you drop the platen? thanks for this.



I've done a lot of experimenting here. Thousands of peices. I landed on a system where 40 and 80 grit are done with the roller any thing beyond is done with the platen down. My experience is that anythihng beyond 80 just cannot remove enough material to remove the chatter marks (scallops) from the previous grit. My sanding regimen aims for no more than one pass at every grit after flattening is acheived. Also, my last grit is always a paper belt while the others are cloth.

[Pat Rice]

As a hobbyist, I am always interested in learning more about things I have never been exposed to. I never have used a wide belt sander and the setup and maintenance sound a little complex. But I do enjoy reading about it.

[Kevin Jenness]

I've done a lot of experimenting here. Thousands of peices. I landed on a system where 40 and 80 grit are done with the roller any thing beyond is done with the platen down. My experience is that anythihng beyond 80 just cannot remove enough material to remove the chatter marks (scallops) from the previous grit. My sanding regimen aims for no more than one pass at every grit after flattening is acheived. Also, my last grit is always a paper belt while the others are cloth.

I could be wrong but that sounds like the drum is out of round. It can be resurfaced if that is the case. Using the platen for calibration passes takes more power and is harder on the belts, and really shouldn't be necessary. I typically step up in grit size no more than 50% - usually 80/120 or 100/150 - and remove the previous scratch pattern in one pass without using the platen.

Here is a chart describing optimum stock removal rates from Surfprep https://surfprep-bucket.s3.amazonaws.com/wp-content/uploads/2020/05/01135208/Wide-Belt-Stock-Removal-Chart-2020-3.pdf. The site has quite a bit of good information on wide belt operation.

[Warren Lake]

I trued the drum and feed on the machine before I used it. I stuck outbody paper onto a straight edge and ran the drum down till it was getting sanded and made it true, Then I ran the track and lowered the head down till it was sanding the track feed to true that. Took minimal off both. Ran the track feed at the slow speed. worked very well for that machine.

[Joe Calhoon]

It does sound like a out of round drum. We had that problem on the old Cemco that had a million miles on it. Going up from 80 you should be able to skip a grit without getting scratches from the previous belt and not using the platen. I rarely use 40 or 60 but you do have to be careful coming up from these to avoid deep scratches.

[johnny means]

johnny - how much do you drop the platen? thanks for this.

It's a felt platen. It only gets dropped about .003 inches below the steel roller. When I switch to 120, I don't adjust height at all and only drop the platen.

[johnny means]

I could be wrong but that sounds like the drum is out of round. It can be resurfaced if that is the case. Using the platen for calibration passes takes more power and is harder on the belts, and really shouldn't be necessary. I typically step up in grit size no more than 50% - usually 80/120 or 100/150 - and remove the previous scratch pattern in one pass without using the platen.

Here is a chart describing optimum stock removal rates from Surfprep https://surfprep-bucket.s3.amazonaws.com/wp-content/uploads/2020/05/01135208/Wide-Belt-Stock-Removal-Chart-2020-3.pdf. The site has quite a bit of good information on wide belt operation.

Why would you think the drum is out of round?

[Kevin Jenness]

Why would you think the drum is out of round?

Because you said you can't get rid of the "chatter marks" or "scallops" from the previous grit without the platen down. "Scallops" indicates to me periodic deeper cuts like what you get from a planer with a fast feed rate, whereas a sanding drum should produce a continuous plane surface if the drum is running true. Maybe there is a problem with the pressure rollers/ bar , or the drum bearings, or the belts you are using (bad splices? kinked from hanging on a narrow rail? swelled splices from wet cleaning?) but what you are experiencing does not sound like the machine is running optimally. If you can make a change so that the platen is not needed for calibration I think you will get substantially longer belt life. Is there a competent technician in your area who could check it out?

[Mark e Kessler]

Because you said you can't get rid of the "chatter marks" or "scallops" from the previous grit without the platen down. "Scallops" indicates to me periodic deeper cuts like what you get from a planer with a fast feed rate, whereas a sanding drum should produce a continuous plane surface if the drum is running true. Maybe there is a problem with the pressure rollers/ bar , or the drum bearings, or the belts you are using (bad splices? kinked from hanging on a narrow rail? swelled splices from wet cleaning?) but what you are experiencing does not sound like the machine is running optimally. If you can make a change so thats the platen is not needed for calibration I think you will get substantially longer belt life. Is there a competent technician in your area who could check it out?


Could be bearings too, if you run at slow speed then a fast speed if the chatter marks are consistently spaced could be a bearing, if the spacing increases/decreases more than likely belt splice, you can also crayon/chalk the splice to see if it transfers the marks on top of the chatter, it’s never worked for me but worth a try. You need a “zero tolerance” splice )i think that is a #1 joint) should be +- 0.001. If you have a steel head like i do (or a very had rubber 85 shore and above) then you will probably not be able to eliminate chatter from a belt. The platen will more than likely never remove chatter ir could infact transfer the belt splice putting that chatter marks in the material if there is too much pressure. I see Chatter from belt splice off my steel head at 120 even but you have to look real close.

[Kevin Jenness]

Yes, it's true that any thickness change at the splice will cause chatter marks and will be accentuated by a hard drum so my statement that a sanding drum should produce a continuous plane surface if the drum is running true is a bit of an exaggeration. A softer drum or platen will give and conform to a splice irregularity and lengthen and soften its effect. 3M uses a finger-jointed butt splice with a very thin tape backing which us probably about as good as it gets, better than a skived and glued joint. Still, having to use the platen to overcome chatter between calibration passes suggests a machine issue.

A couple of excerpts from a discussion on another forum:

Although belt splice causes the most amount of chatter problems, there are other sources as well. The drum could have a flat spot on it or there could be a balance issue with the drum, brake disk and/or top idler roller.

Wide belt chatter can come from a few different things: 1) belt splice, either the type or its change (swelling due to wet cleaning). 2) slippage of the part due to improper hold down pressure or defective or dirty holddowns. 3) slippage in the conveyor drive unit.What you want to do is look for consistency vs. inconsistency. Example: a belt splice chatter will give consistent, equally spaced marks and you may take a crayon and color the splice heavy - then run a piece through and see if the chatter marks are now highlighted with crayon or not. Also, if it is the belt splice, the marks will widen in distance if you speed the feed up or come closer if you slow the feed speed down. Hold down slippage will usually give inconsistent chatter marks and will usually concentrate themselves on the front and trailing ends of piece being sanded. Drive slippage, which can be very consistent due to a broken tooth on a cog or mechanical drive, will usually be wide spaced, in unison with the turning of the drive.
Probably the worst thing to do is start spending big money replacing parts hoping to fix it. This chatter is most often developed at the drum (which by the way can also give consistent marks due to a damage or defect in the drum - but is rare). The crayon will show the difference here. Sometimes this chatter developed at the drum can be eliminated with the platen, simply sanding out the chatter at that trailing stage. First!, learn what is generating the problem, then specifically address that adjustment, repair, cleaning or change in belt.

Johnny seems to have arrived at a tolerable solution so maybe this is academic. I tend toward "If it ain't broke fix it anyway, it could be made better " but that's just me.

[johnny means]

Because you said you can't get rid of the "chatter marks" or "scallops" from the previous grit without the platen down. "Scallops" indicates to me periodic deeper cuts like what you get from a planer with a fast feed rate, whereas a sanding drum should produce a continuous plane surface if the drum is running true. Maybe there is a problem with the pressure rollers/ bar , or the drum bearings, or the belts you are using (bad splices? kinked from hanging on a narrow rail? swelled splices from wet cleaning?) but what you are experiencing does not sound like the machine is running optimally. If you can make a change so that the platen is not needed for calibration I think you will get substantially longer belt life. Is there a competent technician in your area who could check it out?

I don't think you're thinking about the same issue I am. These aren't ridges across the material with undulating thickness like scallops from a planer. It's a very subtle change in the light reflecting character of the wood. The chatter I'm talking about is more like the curl in curly maple. This chatter is caused by the thcker area at the belt seam when it crosses the lowest point of the roller. That momentary increase in thickness isn't changing the scratch pattern, it's crushing the wood fibers beneath it. It's like a knife edged hammer pounding the surface as it goes by. It's almost impossible to sand out this chatter with an orbital, as it goes much deeper than the last scratch marks. Often our trainees will have what appears to be a perfectly sanded surface, only to find a striped, yet, glossy mess when finish is applied. A soft platen gives that hammer knife somewhere to go besides into the material. It also elongates that momentary impact, causing many to overlap, obfuscating the indentations.

[Kevin Jenness]

I don't think you're thinking about the same issue I am. These aren't ridges across the material with undulating thickness like scallops from a planer. It's a very subtle change in the light reflecting character of the wood. The chatter I'm talking about is more like the curl in curly maple. This chatter is caused by the thcker area at the belt seam when it crosses the lowest point of the roller. That momentary increase in thickness isn't changing the scratch pattern, it's crushing the wood fibers beneath it. It's like a knife edged hammer pounding the surface as it goes by. It's almost impossible to sand out this chatter with an orbital, as it goes much deeper than the last scratch marks. Often our trainees will have what appears to be a perfectly sanded surface, only to find a striped, yet, glossy mess when finish is applied. A soft platen gives that hammer knife somewhere to go besides into the material. It also elongates that momentary impact, causing many to overlap, obfuscating the indentations.

If you are happy with the way the system is working so be it. I would be looking at different belts with better lap joints. The belt seam chatter should not be so deep that you have to use the platen at intermediate grits. Free advice, worth every penny.

[johnny means]

If you are happy with the way the system is working so be it. I would be looking at different belts with better lap joints. The belt seam chatter should not be so deep that you have to use the platen at intermediate grits. Free advice, worth every penny.

Believe me, I've had this conversation with the people making the belts and the machines. I'm pretty sure better belts isn't an option. We use cloth for durability. nothing to be done to improve the taped seam. Final pass is a paper belt. using all paper belts is not an option for a sander that runs 5 hours a day. Why wouldn't I use the platen on every grit possible? What's the point of introducing chatter in more steps and not less?

[Kevin Jenness]

Believe me, I've had this conversation with the people making the belts and the machines. I'm pretty sure better belts isn't an option. We use cloth for durability. nothing to be done to improve the taped seam. Final pass is a paper belt. using all paper belts is not an option for a sander that runs 5 hours a day. Why wouldn't I use the platen on every grit possible? What's the point of introducing chatter in more steps and not less?

Because there typically is no need to eliminate the belt- induced chatter until the last pass. Using the platen generates more heat than the drum alone and reduces belt life.