Aluminum base tolerance

Hi guys, I find myself in the need for a bigger aluminum base, right now I’ve been working with a 4 x 7.5 in base.

I’ve talked to several machinists and none of them can guarantee a +/- .001” tolerance.

A CNC machining service tells me they can do it with a +/- .005” tolerance, at best between .002 and .003. Sounds like too much, I don’t know what effects that might have, but I’m guessing, kinda like low spots or areas on my plate.

Why is this? Am I missing something here? Why is it so hard to get to the 1 thousandth tolerance?

Please help me figure this out!

Thanks everyone, I appreciate your suggestions.

Log in to reply   25 replies so far

enrique…. in my view, .001 tolerence is overkill for viturally any purpose other than the very finest of tiny details. Even then it may not actually be required. sure… it’s nice to have a super precise base, but having one that is merely excellent is certainly not going to cause your work product to go to crap. In reality, you probably have more than .001 cummulative movement and play in your press’s bearings… unless your press is like brand new.

Letterpress printers prior to WWII never had anything nearly so acurrate, and they had few problems printing excellent work.

If they can make a .002 to .003 accurate base, I’d say go for it. In the real world, you’ll probably never see the difference. In my shop I used several bases made of local shop machined aluminum, and I never had any problems. I’ve not measured them accurately, but i’d put them a around 5 1000ths or so…. judging from what my machinist told me.

BUT that’s just my opinion. I’m sure the folks who sell super-precise bases will probably tell you otherwise… and that you might not be able to print at all unless you use what they are trying to sell.

You know what Winking Cat? I’m with you on this one, and I’m gonna go for it. The aluminum is going to cost me $45 dollars for a 9x12 in piece, and the machining is going to cost me a whopping $40 bucks, what the hell, it’s not a huge amount of money if I can have a decent printing base.
I will however screen print a grid onto it, and maybe consider anodizing it to give it some extra protection.
At $95 bucks it will be a great little base. I’ll see how much it’ll run me a base for my Vandercook.

Mic6 AL plate comes cast and precision ground. If you’re using standard relief plates, the 7/8” plate should be perfect for you as it comes. If you use deep relief you’ll want to have it machined down some. Looks like $100/sq ft from Mcmaster, but any AL supply house will have this material in any size you need.

The tolerances are within 0.005 on both thickness and flatness.

Just remember that .003 is the thickness of 20# bond. I would think polymer would behave very BADLY with a tolerance band of .003—especially if on the low side.

I would be leery of any professional (ie: I’m paying) machinist who cannot holds a tolerance of .001 on a machined plate.

For best results, the plate should be ground, allowing for even better tolerances, as mentioned by Luke above.

Can someone please help this beginner with this discussion? I have a chase base for my Kelsey 5 x 8 which works fine with polymer plates. But for my C&P 10x15 I would like to buy a base along the same lines as the discussion here. A 7/8 inch is .875 so a bit short of .918. My unexposed plates (Boxcar are .060 KF 152) making the total of .935 which I think would be too thick. What am I missing? Thanks, Neil

You want a base that gets you to type high. In your case, 0.918-0.060 = 0.858.

I would recommend getting a mic6 plate at 0.875 and have it machined or ground down by a competent machine shop.

The alternative, which is way more popular and probably your best choice, is the boxcar base for deep relief. They advertise it at 0.853, which probably accounts for the adhesive backing and can be accommodated with packing. The guys at boxcar are a great resource for both the base and the plates.

If you use ‘standard relief’ plates like the KF95, you should be able to use the 0.875” plate off the shelf.

Type high is .918, so thats what you are shooting forso if your plates are .060 then you need a base .858. i’m sure if you talk to boxcar they can set you straight.

Thanks as always.

Thanks guys, the thing is I always see the term “competent machinist”, but that explains nothing to ome.

The thing here is the equipment, my regular machinist tells me that his machine on that size base (12 in) will hang, and thus he will get a not so tight tolerance.

Other people I’ve talked to tell me they can’t do it because the stock is not magnetic.

These other company I found, they are a ISO 9001 business. And they will CNC machine it.
But its clear now that to get a .001 tolerance machining is not enough, it needs to be ground which adds another $100 dollars at least to the equation. So we’re talking around $200 total.

The McMaster 7/8” stock comes with a +/- .005 tolerance. If that works well, then this +/- .002 tolerance should work too.
Here in Mexico 7/8” stock is not normal and not carried regularly. Shipping it from the US would easily double the cost, then I might as well fork out for the boxcar base.

——— Just to clarify to newcomers ———
The base I need is .875 because: .918 - .43 is .875, that is the photopolymer relief plates I use .40 I think, plus the double sided adhesive I use, and it has worked just fine.

So I already went for it, they are machining it as I write, I will pick it up tomorrow morning and I will let everyone know how good or bad it is.

If it doesn’t work, I can always have it ground to .858 and use the thicker polymer of .60, but I plan to use this base with my Vandercook too.

Thanks everyone for your friendly and insightful comments.

I use a few pieces of machined and grounded aluminum base made by a small machine shop. They don’t have large machines so the have to drill holes at the back of the base, tap them so that the can mount them onto the machine firmly. It also cost me around $100 US to get the job done.

Now the base works fine. It does not work out quite well at the beginning as I did not have a table surface flat enough to lock the form, until I made a workbench with two pieces of surface joined 3/4” plywood as the top and a piece of glass (~5mm thickness). I bet in your area you can find good table top more easily.

have been using .73 polymer plate on a Vandercook press Western with form roller supports-ink was picking up in the bottom of the plate , advice was to trim plate just to text and much deeper polymer plate 1.52……………..I also need to practice!

have been using .73 polymer plate on a Vandercook press Western with form roller supports-ink was picking up in the bottom of the plate , advice was to trim plate just to text and much deeper polymer plate 1.52……………..I also need to practice!


Just a thought. You would not in anyway buy metal type that was not cast at .918 would you? Why then would you even think of taking any risk with a plate/base combo that would be a variance of that?

Perhaps a reminder from Lewis Allen’s Printing with the Handpress?: “inferior tools corrode the spirit.”


This post is senseless if you cant get your head around the comment written by Gerald above . You will never need buy another ,its as basic as buying your press you pay for what you get and if you want good quality you must expect the cost to reflect that. as you said milling alone wont be accurate so you must expect to pay the grinding cost .. a bit like two plus two equals four .

You have to order a thicker aluminum stock, then have that ground down to whatever thickness you want using a “Surface Grind” onto both sides, and not a “Blanchard Grind”. So find a shop that can surface grind.

Or, if you want to get it right, just spend the money and purchase from boxcar. They can cut the bases down to whatever size you want. Because if your shop messes up, and you end up purchasing two or three bases, you’ll end up spending more than what you would have with boxcar. And if you don’t understand how to ask for the correct thing like Peter Luckhurst says, you’re more than likely going to get a base that won’t work.
Good Luck!

One thing which has not been mentioned is that there are a lot of other areas in printing processes which are subject to height variation, and this is why we do makeready to get the print quality we want. The biggest contributor to variation is paper. When I was in the carton industry, there was a 10% caliper tolerance on boxboard. I’m sure other papers have similiar variation. Also, nothing is machined absolutely perfectly, and it only gets worse with use. In a platen press the bed is not absolutely perfect, and neither is the platen. The platen adjustment may be slightly off. The packing and tympan are not perfectly flat either. The lockup might be slightly off. In a cylinder press, the cylinder will not be perfectly round, and the bed will not be perfectly flat if you measure it presisely enough. There will be a T.I.R. (total indicated runout) associated with the cylinder. If you mount a dial indicator on the cylinder carriage with the indicator on the cylinder surface, and rotate the cylinder, you might be quite surprised at the variation present. Or if you mount the indicator on the carriage, with the indicator point on the press bed, and move the carriage along the bed, that would be interesting to see as well. When using photopolymer plates, there is a thickness tolerance on the plate material (when I was in flexo, the photopolymer plate material was +/- half a thousandth of an inch, if I recall). The plates were inaccurate enough that it was not uncommon to check them. The two sided adhesive that you use to stick the photopolymer to the base has a tolerance. And as we are discussing, the plate base has a tolerance.

When you think of all these potential areas of variation, it is amazing we can print at all. We just hope that some of the high places balance out some of the low places most of the time.

So, when we talk about plate mounting base variation, that is only part of the big picture. Is it necessary to have it within .001 inch? Maybe so, maybe not. If one could get a piece of aluminum off the shelf with a .005” tolerance, maybe they should try it first. Remember, if the tolerance is .005, that doesn’t necessarily mean that your piece will be off by .005. That is only the maximum allowable variation. It may be much closer, and whether it is closer or not, it may work just fine. If it doesn’t, it could always be more precisely machined later. If the extra machining makes it slightly low, it could be built up on the back with a piece of mylar or something. But remember, the mylar has a thickness tolerance too (lol).

Geoffrey has it exactly right. It is completely unnecessary to get your base sub thou unless everything else in the system is as well. So, if you have a platen press that has perfect trucks and rails and rollers after all of these years, and your plates and adhesive are also perfect, as is your bed, then maybe it is worth it to get a surface ground plate.

For the rest of us, I would argue that blanchard ground or even a well machined plate will be more than adequate, especially if you’re not in the states where boxcar’s shipping is reasonable.

Geoffrey has some of the most eloquent and straightforward advice I have seen on the topic in 4-5 years of reading this forum.

True enough, and obvious to any seasoned printer, but isn’t the task to eliminate variance rather than ignore it? or worse, add to it. There is a point of reference to aim for, disregard this at one’s own peril, unless one doesn’t really care.

The difference between fine printing and everything else is a thousands of an inch. Isn’t that why we have makeready, overlays, interlays, underlays, etc.?


I want to take the opportunity now that this discussion has resurfaced a bit. I did make the base. And I was there at the shop with the engineer to witness personally the tolerance the base had at the end.
He used a Mitutoyo comparator to measure first the variation in height, and I saw it with my own eyes +/- 0.001”, then for flatness tolerance we got +/- 0.0005”.

So I took the risk but it was worth it. The idea here wasn’t necessarily to go frugal, it was my solution because I had an urgent job I had not planned for and the base I already had was too small. There wasn’t enough time to get a boxcar base shipped to Mexico, and I didn’t want it to be super expensive either in case it didn’t work well.

Already used it, printed very well, and I am happy with the results. The base was not ground, anodized or screen printed (grid), but it got the job done.

Probably these specifications would be hard to replicate, maybe even with the same company, and I don’t necessarily recommend anyone that they take the chance of not getting an accurate enough base machined.

And to finalize: I profoundly appreciate your thoughts, and I believe firmly in the pursue of eliminating of variables in the printing equation.

Gerald, you are right. My main purpose in my post above was to make people aware that there are a lot more areas of potential variation than their plate bases. I think some people, especially the less experienced ones, tend to concentrate on one thing and do not always take into account the big picture. We were all “less experienced” at one time, and one of the reasons I post on this site is to help people move past that category.

HavenPress, thank you for the complement.


Yes, but will they? Older generation printers were either trained or learned through self-directed study. At a time when there was no Easy Button to click on. The majority of new entries into letterpress get their introduction through workshops or short courses, and that’s it. They have no idea of what the big picture is. Do internet forums really help in this regard?—when every quick answer to every quick question is given the same weight regardless of the experience of the respondent, and especially so on forums where identities can be concealed.



There’s not a lot of opportunities to make a career out of letterpress these days; its survival lies in the hands of the hobbyists. The new entries into letterpress are the future of letterpress.

There will always be the folks that want the “easy button” route to success — that isn’t unique to letterpress. Go on a metalworking forum, and you’ll see the “just bought a lathe on craigslist, can someone tell me how to build a motorbike” folks. The good news is that they’re not a permanent fixture. Those people will eventually come to learn that it isn’t easy, give up, and move on to the next shiny thing.

Forums (and mailing lists) are valuable. For those who are committed, having a place to pose questions is important. I don’t have anyone that I can ask, online communities are all I have. A solution to the problem of throwaway questions? Don’t answer them when they come from the obviously lazy or the willingly uninformed. Often a quick peek at the poster’s history will paint a clear picture.

As for anonymity; it bugs me too. We’re all adults here and should communicate like adults. I post on a first name basis only, my full name guarantees an instant match on any search engine. I had an experience involving being Googled by a Canada Border Services agent while crossing the border that made me realize how tenuous our privacy is. In any private communication, I’m quite open with my identity.

- Keelan

Hi Keelan

Thank you for sending the private post.

Yes, obviously new entries are the future of letterpress. I think it best to guide them though rather than encourage them with dubious and ultimately frustrating information, or yes, as you suggest, they will go away, and do.

I teach letterpress printing and many of my students have gone on to establish themselves as fine press book printers, commercial letterpress printers, private presses, etc. So I wouldn’t agree that the future lies in the hands of the hobbyists. Not to discount them, but opportunities are out there. However, one has to prepare for them. It’s not easy by any means but it can be done.

I met Harold Kyle back in 2000 when I was giving a Vandercook maintenance workshop at Minnesota Center for the Book Arts. He had worked for Campbell-Logan Bindery, had his own print shop, and was then serving a residency at the Center and fooling around with this idea he had for a new type of flatbase. I was encouraging. Boxcar Press now has over 90 employees and dominates the letterpress photopolymer plate market.

While forums can be useful for the occasional bits and pieces—I own and moderate a couple of them—they are no substitute for dedicated research and education, guided training, experience…


Then again, newbies can always post to Letpress or Ladies of Letterpress, though they would have to adjust to the environment (remember what high school was like?)