Pressure Transducers

I have an application where I will have a tonnage cylinder and a jack cylinder. The tonnage cylinder will single ended and have a 14" bore. The jack cylinder will have a 5" bore and a 3" rod. This would be approx a 19" bore and a 3" rod. I can put this into RMC Tools and it seems to scale correctly. Will I have trouble with this?

I don’t understand.

  1. Normally there are two jack cylinders and one tonnage cylinder or one jack cylinder and two tonnage cylinders so the system is symmetrical and it doesn’t bind. You must have some mechanical way of keeping the cylinders from binding.
  2. Are you saying the tonnage cylinder is a ram? A ram is where there isn’t a rod just a big slug of metal with the diameter as big as the cylinder.
  3. The areas are additive, not the diameters. The area of the to pistons/rams is
TotalCapArea=(PI/4)*(14^2+5^2)=173.573 sq in.
TotalRodArea=(PI/4)*(5^2-3^2)=12.566 sq in.
EquivCapDia=sqrt(14^2+5^2)=14.8666 inch diameter.
EquivRodDia=sqrt(14^2+3^2)=14.318

The equivalent rod and cap diameters must be entered into the RMCTools scaling wizard. This assumes the tonnage cylinder is a ram, where the rod is effectively the same diameter as the cylinder, and that the jack cylinder will have pressure feedback to measure the opposing force. Given my assumptions are correct there should be no problems.

This would make a good Friday math quiz.

The jack cylinder is mounted on top of the tonnage cylinder.

Correct it is a ram.

True I was thinking simply. Also the rod of the jack must be subtracted from the area of the “ram”

That’s what I thought.

One other thing. The jack cylinder will be in the circuit all the time. The ram will not. So I am planning on putting the transducers on the jack cylinder. When the ram is out of the circuit my pressure scaling will be off. Should not hurt anything in the process but it could give me some odd numbers for tonnage on the HMI.

Should I setup dual scales for tonnage feedback? One for the jack and one for both.

In previous systems the tonnage cylinders were open to a fill tank using a fill valve. This way the servo valve only needed to supply oil to the jack cylinders when the press was closing. This allowed for faster closing speeds. Eventually the fill valve must be closed and the servo valve pressurizes the jack and tonnage cylinders. This should be done just before the press closes and starts to pressurize. If the fill valve closes too early then time is wasted as the press closes at a slower speed. If the fill valve closes after the jack cylinder starts to pressurize then the jack cylinder will depressurize into the tonnage cylinder. This just doesn’t look good on the graph. The user program makes it easy to enable an output to close the fill valve and to have the servo valve fill both the jack and the tonnage cylinders. This can be done when a position is reached and/or a pressure condition is met.

The picture shows a press with on jack cylinder and two tonnage cylinders or two jack cylinders and one tonnage cylinder. It has been awhile so I have forgotten. One can see the press closing at high speed until about 10 seconds. At that point the press was close to being closed. At this point we closed the fill valve and slowed down the press closing speed because now the servo valve had to fill both the jack and the tonnage cylinders. At about 13 seconds the press is finally closed and pressurizing. This takes a while because the servo valve must pressurize the tonnage cylinder. One can see it takes considerable control output ( green ) to pressurize the tonnage cylinder. At 1500 psi the press switched to pressure control mode where the pressure PID takes over. Target pressure was ramped up nicely using s curves to 2000 psi. Notice there is little error between the target and actual pressure. Most of the control output came from the pressure feed forwards. As far a I know we are the only ones that offer feed forwards for pressure. After a couple seconds at 2000 psi the pressure was reduced slowly. In this application the pressure couldn’t be released quickly because and trapped air bubbles would explode and cause little pits in them material After depressurizing the pressure was applied again. This removed the air from the material. Anywhere from 2 to 7 ‘bump’ cycles were used.

All the sequecing was controlled using the RMC100’s step tables.

What are some of the brands and models of transducers that work best. I have a customer the is spec <1ms rise time and .1%fs and manifold mount. Can you get close.

I just added a topic for pressure transducers: Pressure Transducers

The Bosch HM18 and Ashcroft G2 look pretty good.