In this podcast, Tom Rohlfs talks with John Maher about the plastic manifold design process. He explains what customers should expect when they work with a manifold manufacturer through the design process.
John Maher: Hi, I’m John Maher. 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 plastic manifold design. Welcome, Tom.
Tom Rohlfs: Thanks for having me.
John: Sure. So Tom, do customers typically come to you with a design already planned out for their manifold, or are they looking for you to design it for them?
Tom: We have usually two requests. I get involved in design discussions almost on every new project. Customers don’t often know what they can and can’t do. From the design of a manifold, where are the boundaries? Where does it make it difficult to machine? Where does it make it difficult to produce? They need support on that.
Additionally, materials as well, materials selection, material choice can be a little tricky, especially because there’s many, many different chemicals and reagents out there. They need to know that their chemistry works with a particular material and they’re looking for my support on that. So generally speaking, it’s a support at the design level where they have their design layout and they need us to review it. Secondly, they also have customers who come to us who are looking for a full design. When I say a full design, I mean they’ve worked out the chemistry. They have a basic schematic. They breadboard their process internally, they see that it works.
They’re at the point where they want to turn it into a working manifold, moving closer to more of an alpha or beta stage for their device. That’s where we step in and we can support them on building out the entire manifold from material choice, size, what components they’re going to integrate into it.
Most often at that juncture, they know which components they want to use. They may need help selecting valves and pumps and things like that, but they have a really good idea of where they’re going with it. So he had those two points. One where the customer’s done essentially their own internal design. They’re looking for design review or alternatively a full design, which we would do for them, or we’re helping select various components and really figuring out the entire space of the bonded manifold.
John: Okay, and what are some typical requests that people want in a manifold?
Tom: We have a conference room at Controlled Fluidics, and it’s loaded with manifolds of all different shapes, sizes. They have an infinite number of requests. What are the basics? Of course, what do we see is kind of a repeating pattern? Obviously every manifold needs a valve, right? You need to control your fluid, be it in out which branches it goes to. Every manifold has a valve.
Oftentimes pumps are a common request, pressure sensors, temperature sensors, anything that’s going to control their process. I’d say the workhorse of a manifold is the valve though, the valve really controls that fluid flow. Pumping can come from other places, pressure relief discs. It’s really quite a bit of different applications used widely with different manifolds.
One thing that’s interesting is some of our customers and what we’re capable of, which is unique to Controlled Fluidics, is we can integrate reservoirs or accumulators into our manifolds, which is basically just an open space. Oftentimes, historically, a reservoir or accumulator was separate to your fluid flow as its own device, its own discrete device.
What’s nice about our abilities to control fluidics is we can cut that reservoir right into the manifold. So you’ll have a big pocket. We have an aerospace application where it’s a three layer manifold where it has two accumulators in the back, and then on top of it’s the channel selection. So that’s something that’s unique to Controlled Fluidics. We’re very good at building those types of features into a manifold.
John: What are the reservoirs used for? What process is that useful for?
Tom: From an electrical perspective, it’d be considered a capacitor, right? It’s just a reservoir to hold a certain amount of pressure, a certain amount of liquid for their process, and then maybe they pull from that reservoir or they dump back into it as part of their process.
But you see it a lot in pneumatics where they want to contain some amount of gas inside that manifold for… for example, we have an aerospace application where it’s a test device for planes. When a plane lands, they have a test device, which they hook up to all the plane sensors and they need to test the sensors. Well, the sensors in a plane need some amount of working air to verify that they in fact work properly.
So the reservoir in the manifold provides that working fluid to the plane sensors so they can verify the sensors are in fact still operating. And I believe that type of testing happens every time a plane leads. Obviously, everybody wants the sensors on a plane to work effectively when they go up. And so that’s part of that whole landing procedure. The plane lands, they run over, they check out all the sensors, make certain everything’s good to go for the next flight out.
John: Okay. You mentioned before that sometimes a customer will come to you with a certain design, but they don’t really maybe understand what’s easy or very difficult for you to produce. What are some of the things that are a little bit more difficult for you to produce that a customer might come to you and then you say, “yeah, we can’t really do that”?
Tom: Sure. So customers come, oftentimes with questions about a scale. They want to know, “okay, my manifold is this big, is this within your working capacity?” We talked about that 12 by 18. So that probably fills 99% of the manifold universe for customer requests. But oftentimes it’s, “can I make this feature?”
Manifolds have gotten quite complex and channels within the manifold can run very close to each other. And so the biggest risk of bonded manifolds, well, what we would consider a failure point is when one channel leaks to the other, we call crosstalk, right? And so the closer you put those channels together, the more likely they’re going to crosstalk, well, what is that limit? How close can I compact all my channels before I’m going to risk that crosstalk? So for us, our standard is one millimeter. We like to keep every feature away from other features within a manifold channel layout within a millimeter.
So we consider the web thickness to be strong enough. And crosstalk can occur because of pressure. Right? You have pressure in one channel and not in another. And then if the bond isn’t adequate, the pressure leaks from one channel to the other. So we found that our manifolds with that one millimeter spacing can run consistently up to 150 PSI without any problem, which is quite a bit for an acrylic manifold.
Now interestingly enough, we have some customers running Ulta manifolds. They go up to 500 psi, and one unusual application that almost runs there is over a 1000 PSI within a plastic manifold. Very small channels, of course, a fairly simple manifold, but still, our bond strengths are so strong that we have capabilities up to a 1000 PSI in unique applications in an internal manifold. Bulk of the applications are under a 100 psi, but our bond strengths are so robust with that 1 millimeter spacing that you can run quite high pressures, which is surprising. One wouldn’t expect it in a plastic component for sure.
John: And when you have that kind of high pressure, do you have to have the channels be a little bit further apart in order to accommodate that?
Tom: Yeah, it’s size. If you’re putting a 1000 PSI in an eight inch channel, that wouldn’t be good. So these are very small half millimeter channels, so there’s not a lot of force internally. But nonetheless, a 1000 psi. Again, unusual application. Typically we would warranty our manifolds up to 150 psi for essentially any configuration. So even that, our bonds strengths are quite good.
So that’s one of the things that people like to talk about is “I’m doing this layout, I want to put these features here. This one’s coming up near an edge, this is coming next to another channel. What kind of spacing can I do? What sizes can I do?”
Another thing we’re finding a lot of demand for is optical windows. Customers want to do some sort of imaging. We have applications where they’re cell sorters and they want to be able to count cells through a manifold bond. A manifold can be a simple cell sorter where you put a cell in one side, and it goes through a viewing window and comes out the other. They want to be able to see it accurately. Other customers want to shine lasers through them. Sometimes they shine UV lights through them. So we’re finding that the requests for optically clear portions of the manifold have really come up and they have a specific side here, this is the spot I want, can you do it? So we talk a lot about that as well.
John: So how do you actually go about doing the manifold design, actually doing the design itself, and how do you present that design to the customer before you start the manufacturing process?
Tom: Sure. We’ll work for their engineers. We need to talk about layout, physical space, what physical space is this going to fit in? What kind of layout are you looking for? Material choice. We’ll work closely with the engineer, be it on the full design or even design support. And we’re kind of through those important touch points so that we know we can manufacture it.
We talk about complexity, cost, the components that they want to mount on it. Generally speaking, once we gather all that information, we’ll do a layout, a solid model layout where they can, then we present that to them. Oftentimes they get one round of revisions. So we’ll present our idea, the customer will review it and decide if they want to make changes or they like it. They get one round of revisions, we return it back for approval. Once it’s all approved, we’ll go ahead and start the manufacturing process.
John: So during that process where they’re reviewing it, you’re actually making a physical manifold product for them to review, is that correct?
Tom: No, I’m sorry. When I said solid model, it’s a CAD model.
John: I see.
Tom: But it’s a 3D CAD model that you can see, you can visualize in space quite easily. Shows you can rotate, see all the sides of it. A solid model’s a solid CAD model.
John: Right. Okay. And then how do you choose materials? Is that part of the design process, figuring out what the best material is to use?
Tom: Definitely. That’s probably the number one question everyone has. We have a fairly extensive library about chemical resistances to plastics. And so everyone has, oftentimes I’ll get a list of 5, 10, 15 different chemicals that this manifold is going to be exposed to. And I’ll pick through my chemical resistance guides to make certain that we select the right plastic.
The downside to plastic is that it’s stress sensitive. Some of these reagents can attack the plastic because they might be solvent based, and then that plastic ultimately will fail. So we have to pick the right plastic that can chemically resist all the reagents the customer’s planning to use.
John: And then what are the next steps after a design is complete, how do you then take it to the next level?
Tom: So the next level is after design approval. Generally we’re going to create a prototype for the customer. We can either 3D print a simple model, or we can do actual pre-production run and generally it’ll be five to 10 pieces, and that’ll allow the customer to actually test it out with their particular reagents and make sure that their processes are working correctly before we were to move to a production type of environment production run.
John: All right. Well, that’s really great information, Tom. Thanks again for speaking with me today.
Tom: Yeah, you’re welcome.
John: And for more information, you can visit the website at controlledfluidics.com or call (603) 673-4323.