Common Concerns of Bonded Manifolds (Podcast)

John Maher: Hi. I'm John Maher, and I'm here today with Tom Rohlfs, President and Principal Engineer at Controlled Fluidics, a plastics machining company specializing in precision manifolds. Our topic today is common concerns of bonded manifolds.

Welcome, Tom.

Tom Rohlfs: Hello, John. Thanks for having me.

Reducing Stress-Cracking Risk With Bonded Manifolds

John: Yeah. Tom, we're going to talk about four different aspects that are concerns that people might have about bonded manifolds. They are stress cracking, cost, threading, and material choice. Let's just go through each one of those and you can tell us a little bit more about them and why they're something that people should be considering when they're looking into manifolds.

The first one was stress cracking. Talk a little bit about that and what it is and how that plays into how a company is going to design and make a manifold for their application.

Tom: Yeah. Sure. Glad to help. Of the four topics, I'm going to combine stress cracking with material choice because they really go hand in hand. Oftentimes, when a customer calls me, they're looking for support on bonded manifolds. They've got a design in mind. They've got a sense of their process. They've got, "Hey, I know I've got, I've bread-boarded it, right? And bread-boarding means they have a bunch of valves and a pump and tubing and they've tied everything together. They've run their chemicals through it and had a wonderful result.

And they want to say, "Hey, okay, we're going to go into production with our piece of equipment. We want to boil all this tubing and discrete components down into one manifold. And we're worried about things. We don't know everything about a manifold because we've never built one before." And so their questions tend to really tie into, say, those three topics with material choice and stress cracking as being combined as one. Because stress cracking by the sound of it isn't a good thing, right? It's a failure of manifold and the customers obviously want to avoid that.

John: Mm-hmm.

Choosing Materials for Bonded Manifolds

Tom: So the first question right off the bat is, "What materials should I use?" And a lot of that just goes back to, I have a few basic questions. Manifolds themselves are normally not a structural component, so we don't worry so much about strength. Nobody's going to bolt a large portion of their equipment to it and expect it to kind of hold up the weight of the equipment. It isn't a structural member, it is typically just a control device of either pneumatics or fluidics.

So generally right out of the box, that's the first question, "Are we're talking about a pneumatic manifold or a fluidic manifold?" As soon as the customer can identify that, it starts narrowing the number of choices. Amorphous materials are great, and these are the materials that we use because they can be bonded. If you have a semi crystalline or a crystalline material like PEEK, it isn't readily bonded through regular thermal type approaches.

So the amorphous materials allow us to do what we do, build multilayered manifolds. But unfortunately, because they're amorphous, they're also sensitive to stress cracking. And stress cracking is when either the thermal, mechanical or chemical stresses, the chains of the amorphous material are broken and it actually cleaves the material.

Generally, if you look at a part that's been stress cracked over time from some attack, it can look like fine spider webs across the surface, or you can actually see entire cracks running through the part. It can get so severe that actually a piece of plastic can fall out. So it can cleave right through the part in say a corner and fall right out.

So picking the right material for those particular applications is always critical. So what we want to do is focus on do you have any thermal constraints? If you have thermal constraints where you need high temperature resistance because you're running hot liquids or hot gasses, we're going to be looking at the Ultems or the Polycarbonates or the COP.

If you have just simply a room temperature part, we're looking at, that's running pneumatics, we're looking at acrylic, because why would you want to use acrylic? It's low cost. It's the lowest price point for a manifold. So we look at that side. So material choice and figuring out what the customer needs often starts with chemical exposure and thermal concerns. And then beyond that, again, we don't dip into tensile strength, but all these efforts are to essentially avoid stress cracking failures.

If you marry the wrong material like acrylic with an aggressive chemical like acetone, the manifold will simply stress crack and fall apart and break. So we want to be very careful. We have significant chemical resistance guides. Oftentimes the customer will say, "Well, I'm running this chemical, this chemical, this chemical." I'll dive into my chemical resistance guides and can provide a recommendation based on results from those guides on whether which material is appropriate or not.

John: Okay.

Tom: So that's what material choice is all about, really narrowing down which material fits the application and our knowledge in that is pretty extensive. We have many, many catalogs that detail chemical resistance, and again, the focus is on long-term viability of that material in particular.

Where Do Stress Cracks Occur in Bonded Manifolds?

John: So both thermal issues and chemical issues can cause stress cracking. In order to make a bonded manifold, you're taking two pieces of plastic and you're bonding them together. Do the stress cracks tend to happen in that bond between the two pieces of material, or can they happen anywhere in the manifold?

Tom: They can happen generally in the joint. It's a good question. Generally in the joint we don't see stress cracking. Our process is robust and well-defined. We understand the limits of the plastic and we stay well within the bounds so that our process itself does not cause cracking.

However, what you do see is oftentimes if a chemical has been flown that's too aggressive, it's gone through the channels, you'll see stress cracking at the channels. Or interestingly enough, and I guess I should have spoken to this sooner, is that every input and output in a bonded manifold is a threaded port or some type of port. Very common for a customer to over-torque their fittings into the manifold and have cracking right around the fittings. It happens all the time.

Getting the right torques for a particular fitting size is really critical or else long-term steady state stress is the thing that causes that. This is from a tensile stress perspective actually, long-term steady state stress and over tightened fitting is a very common fail point. And in that regard, it's oftentimes a customer education effort. We offer a torque chart for every different size fittings that we could share with our customer to help them assemble their manifold properly.

How Pressure Affects Stress Cracking

John: Does the pressure or PSI of the material being flown through the manifold, does that affect stress cracking as well or can that cause stress cracking high pressures?

Tom: It could, yeah, yeah, sure. And it definitely could because it's a stressor, right? It's a tensile stresser. It definitely could. Generally speaking though manifolds, 90% of the manifolds run under a hundred PSI. Our manifolds easily can tolerate 150 PSI or more. We have actually some customers running four to 500 PSI successfully and not having the manifolds fail.

So generally speaking, the manifolds are much, much stronger than the working pressure that our customers typically use. So it isn't something that we run into all that often.

Potential Threading Issues With Bonded Manifolds

John: And then a threading was a common concern. Is that what you were talking about with attaching hoses and things like that onto the sides of the manifold? Is that what you call threading?

Tom: Yep, exactly.

John: Okay.

Tom: So there's two types of threads here that we would like to kind of focus on. It's one that you just referred to as ports. A very, very common port is an upchurch 1/4-28 flat-bottom port. Generally speaking, customers put in a soft fitting to interface with that. It's hard to over-tighten those, but it happens, especially if they happen to choose a metal fitting instead.

One that does, it can be a gotcha for our customers that we don't necessarily like them, but some customers like to use pipe threads. Pipe thread is a tapered thread, so it's got a wedge type of effect. And its sealing occurs because pipe thread has a taper to it, and the sealing comes as you tighten it that taper wedges tighter, tighter into the thread. It seals great, but it's a plastic manifold. So you could imagine tightening a pipe thread into a plastic puts it under some stress, and over time that can cause stress cracking.

Interestingly enough, a majority of our applications with pipe threads are successful. I think it's because customers don't handle the threads and are very careful with them. But that's the type of thread that can cause problems in the long run, it's that tapered effect on it.

The second threading situation, this conversation comes up all the time, is when customers are attaching valves to the manifold. Generally speaking, we see manifolds with valves of 50, a hundred valves on it, a huge amount, and they're all stacked very closely together. The question becomes, I'm screwing the valve onto the manifold. A lot of times they're very small screws. Number two or smaller, oftentimes even odd 80 threads. These are tiny little threads going into plastic. Customers are worried that if they were to over tighten their valve when assembling it, they strip the threads. Soon as you strip the thread, you've ruined the entire manifold.

It's a legitimate concern. So what we like to do for number two threads or down or two millimeters or down in metric, we like to put in an insert. So what we'll do is we drill the hole and we buy what's called a stainless steel insert, and we thermally stake a pen what's called a Pemco insert into the plastic manifold. So that instead of screwing your odd-80 thread into a small plastic thread, you're screwing it into a piece of metal, it has a lot more robust about it so that if you ever have to service that valve, you want to take it on and off. It can handle almost a reasonable amount of installations and de-installations without stripping. Of course, if it was a straight plastic thread, it would ultimately fail because it's just too weak.

How Cost Affects Material Selection

John: Right. Okay. And then the final concern was cost. How does cost factor into, again, material choice and that choice of what type of plastic you use in your manifold?

Tom: Manifolds. Body manifolds are not inexpensive. Cost of a four-by-six two-layer acrylic manifold with medium density of channels is often about a hundred pieces. It's often about $200 a piece. So if you want to buy a hundred pieces, you're going to pay about $200 a piece, quite a bit of money.

That can be a little shocking for customers. But if they look at total cost of ownership when it comes to having discrete components and tubing and the possibility of leak points and servicing, ultimately the body manifold is the better choice.

As you add layers for acrylic manifolds, you typically add about a hundred dollars per layer. So a three-layer manifold now is $300 and so on and so forth. We don't see manifolds that get much beyond five or six layers because what we find is the risk of failure, the fallout concerns outweighs the cost of having two manifolds, taking six layer manifolds and breaking it into two smaller separate manifolds is generally more cost efficient than trying to buy build one very thick one.

We just have too much fallout, too much risk because every part, a six layer manifold is actually seven pieces. Every layer is a part, and then we bond it together, and then we do the final machining on the entire assembly. So there's a lot of risk there. And if you have too much fallout, it gets very, very expensive.

We can lose up to 10% of our parts in set up and adjustments. They are precision devices, and we have to be accurate. And so there can be that much followup. And so we find that, generally speaking, if you get too many layers, we prefer our customers to break into two, just to help on that cost, especially on the servicing side. If you lose one manifold for some reason, now you haven't lost both. We only lost half the problems. We see that as an advantage as well.

So there is a limitation on all that, and we think that that's kind of where things make the most sense and where that cost becomes more.

On additional materials. We talked earlier that polycarbonate is 10 to 20% more expensive. Ultem can be significantly more expensive. Ultem prices are quite a bit high. COP is someplace in the middle.

So on the higher performance manifolds, material which simply costs more. But that just gives you an idea of where we start. We always try to start with the lowest cost solution for our customers. We could provide them with an acrylic manifold. That's what we want to sell them. We don't want to sell them an Ultem manifold. We're an acrylic now. We try to be very cost-efficient for our customers. We want their product to be successful.

How Do Single-Use Needs Affect Cost?

John: And we've talked before about single use applications as well, where you might have a customer who only uses a manifold once or something like that, and then they have to replace it with another piece. So cost obviously becomes a factor where if you're paying $200 for each manifold, that's going to be much more expensive if you're using a different manifold every single time versus being able to reuse it. How does that sort of factor into the cost as well?

Tom: Yeah, I have an interesting story. So you're spot on there in the consumable, where consumables are required and they need to be a manifold. We look to injection molded solutions for very simple manifolds. We can provide injection molded components and bond them together. That drives the cost way down into the sub $20 range. So that's a possibility for high volume consumable type applications.

I was smiling when you said that because we have one customer who's consumable. I always get tricked. Sometimes I've priced a manifold down and it's two or $300, and I think, "Oh, this is something they'll use for a long time." They announce it's a consumable, and I say, "Well, how much does the total system cost?" They say, "Oh, we're in pharmaceuticals. Our consumables are over a hundred thousand dollars a piece."

John: Wow.

Tom: One small portion of it. So I always get surprised. I think consumable always means a $5 part, when in fact, in some markets in the pharmaceutical industry, they're way up there.

John: Right, so we have to keep that in mind that that $200 for us might sound like a lot and for them, it's like, "Yeah, we'll just throw it away."

Tom: Yeah, yeah, they're, they're not worried about it. Some suddenly saw high-end pharmaceutical type, it's drug development, and some of these drugs are very expensive. They require extremely high purity and very, very precise control of their liquids and gases make the particular drug. And so for them, having consumable at that level is makes economic sense. Hard to imagine, but it's the truth.

Contact Controlled Fluidics Today

John: Right. All right. Well that's really great information, Tom. Thanks again for speaking with me today.

Tom: You're welcome. Take care.

John: And for more information, visit the website at controlledfluidics.com or call 603/673-4323.