Infoworks - Conduit Volume and Preissmann Slot Determination

[samglo]

I'm working on models which include large diameter storage tunnels which fill with open channel flow but can also experience a significant amount of surcharge. I'm seeing more artificial volume being created than what I would expect to see based on the Preissmann slot width. I've tried to minimize the Preissmann slot volume with a small minimum width and celerity ratio of 14.414. For a diameter of 24 feet, I'd expect a slot width of 0.23 feet. For larger surcharging, this width matches my observations for the additional modeled volume in the conduit. For smaller amounts of surcharge, I'm seeing an additional increase in storage volume beyond what can be accounted for with the slot width.

Is the width calculated as shown below (from http://blog.innovyze.com/wp-content/...ICM_and_CS.pdf)

Capture.JPG

or is there an additional transition at the top of the conduit which creates additional volume? Is there a relationship for this?

Thank you,
Sam Glovick

[Robert Dickinson]

Hello Sam,
Have tried the Preissmaann slot and baseflow correction as described in the ICM help file?
Regards,
Robert Dickinson

[Duncan Kitts]

What you may be seeing is what is described in the help page 'Hydraulic Theory' as the 'transition region...included within the model defined by a monotonic cubic between the true pipe geometry and the width of the Preissmann slot'.

The way this works, is the simulation parameter geometry_table_entries (no of geometry table entries’ defines the number of equally vertically spaced points in the geometry table for a pipe from its invert to its soffit. By default, this is 15. The pre-processor calculates width and area for each table entry using the shape’s geometry up to the last but one point. It then fits a monotonic cubic equation for width, given the width at points (geometry_table_entries – 2), (geometry_table_entries – 1), and the slot width. The top of the slot transition is two vertical space steps above the soffit.

Thus, if the vertical space step, deld = height / (geometry_table_entries – 1)

Depth Point # Geometry
Height -deld geometry_table_entries – 1 True
Height geometry_table_entries Transition
Height + deld geometry_table_entries + 1 Transition
Height + 2 * deld geometry_table_entries + 2 Slot

[samglo]

That is perfect! Thank you for the detailed description.

How advanced is the Preissmann slot correction? Does it account for that transition or does it subtract the slot width x pipe length from the specified upstream/downstream shaft area for the entire shaft height?

[Duncan Kitts]

Numerical correction applies additional storage to compensate for three elements, the Preissmann slot, the transition between the slot and the conduit and the base flow depth.

[Andy Bolden]

What you may be seeing is what is described in the help page 'Hydraulic Theory' as the 'transition region...included within the model defined by a monotonic cubic between the true pipe geometry and the width of the Preissmann slot'.

The way this works, is the simulation parameter geometry_table_entries (no of geometry table entries’ defines the number of equally vertically spaced points in the geometry table for a pipe from its invert to its soffit. By default, this is 15. The pre-processor calculates width and area for each table entry using the shape’s geometry up to the last but one point. It then fits a monotonic cubic equation for width, given the width at points (geometry_table_entries – 2), (geometry_table_entries – 1), and the slot width. The top of the slot transition is two vertical space steps above the soffit.

Thus, if the vertical space step, deld = height / (geometry_table_entries – 1)

Depth Point # Geometry
Height -deld geometry_table_entries – 1 True
Height geometry_table_entries Transition
Height + deld geometry_table_entries + 1 Transition
Height + 2 * deld geometry_table_entries + 2 Slot

When viewing the information which is placed in a Notes for a conduit when Preissmann compensation is applied, it includes a compensation volume. I have separately corresponded with Innovyze Support because it appears that the volume is not calculated in the way one would expect. For the model I was working with the ICM calculated slot compensation was somewhat different from hand calculation. I did several cross checks, a precise calculation of the geometric volume of tunnel and shafts and filling the tunnel slowly to see how much volume was taken in. I also checked with a flatter longitudinal profile than actual so as to all-but eliminate the effect of the gradient on the slot volume. I could not match ICM's calculation - it was smaller.

Is there a diagram which shows how the volume calculation is undertaken and also includes the detail of the transition shape from pipe into slot?
The ICM help says that compensation can be negative but the minimum plan area at nodes (in the Simulation Parameters) e.g. 2m2 ensures it can never fall to zero.

The help doesn't explain what happens to the difference, i.e. if Actual Plan Area - compensation area < 2. What if the result of this calculation is (say) -50m2, does ICM ignore 52m2 of area?

When I compared the slot plan area with the applied compensation area (applied to upstream node only) they were rather different. This led me to wonder how the compensation volume is calculated in ICM. For shallow graded pipes it is predominately a plan area compensation. The volume in the slot only becomes significant as the gradient increases.

There are pitfalls with applying slot compensation storage so, notwithstanding the above doubts, it pays to check that it is being applied correctly

• you may have a shaft on a tunnel which extends almost to ground level. Being mindful of defaults you may set the 'chamber roof' at cover slab level rather then leave it at soffit (and perhaps reduce the ICM 'shaft' to the plan area of the man access). It may be that there are high level pipes coming into the tunnel so these might affect the default chamber roof level. If you then apply slot compensation to the ICM 'shaft' it may never be invoked as the water level may not reach that high! A better approach would be to set chamber roof at the soffit of the tunnel and set the ICM 'shaft' area = 'chamber' area.
• If your tunnel shaft is modelled as a storage node, slot compensation does not get applied to it (no warning messages!)

[Duncan Kitts]

When viewing the information which is placed in a Notes for a conduit when Preissmann compensation is applied, it includes a compensation volume. I have separately corresponded with Innovyze Support because it appears that the volume is not calculated in the way one would expect. For the model I was working with the ICM calculated slot compensation was somewhat different from hand calculation. I did several cross checks, a precise calculation of the geometric volume of tunnel and shafts and filling the tunnel slowly to see how much volume was taken in. I also checked with a flatter longitudinal profile than actual so as to all-but eliminate the effect of the gradient on the slot volume. I could not match ICM's calculation - it was smaller.

Is there a diagram which shows how the volume calculation is undertaken and also includes the detail of the transition shape from pipe into slot?
The ICM help says that compensation can be negative but the minimum plan area at nodes (in the Simulation Parameters) e.g. 2m2 ensures it can never fall to zero.

The help doesn't explain what happens to the difference, i.e. if Actual Plan Area - compensation area < 2. What if the result of this calculation is (say) -50m2, does ICM ignore 52m2 of area?

When I compared the slot plan area with the applied compensation area (applied to upstream node only) they were rather different. This led me to wonder how the compensation volume is calculated in ICM. For shallow graded pipes it is predominately a plan area compensation. The volume in the slot only becomes significant as the gradient increases.

There are pitfalls with applying slot compensation storage so, notwithstanding the above doubts, it pays to check that it is being applied correctly

• you may have a shaft on a tunnel which extends almost to ground level. Being mindful of defaults you may set the 'chamber roof' at cover slab level rather then leave it at soffit (and perhaps reduce the ICM 'shaft' to the plan area of the man access). It may be that there are high level pipes coming into the tunnel so these might affect the default chamber roof level. If you then apply slot compensation to the ICM 'shaft' it may never be invoked as the water level may not reach that high! A better approach would be to set chamber roof at the soffit of the tunnel and set the ICM 'shaft' area = 'chamber' area.
• If your tunnel shaft is modelled as a storage node, slot compensation does not get applied to it (no warning messages!)

As per the help topic 'Preissmann Slot and Baseflow Correction':-

"Because the Preissmann Slot correction involves reducing the plan area of the manhole, it is possible for the sum of all storage compensation parameters to be negative. The Minimum Plan Area at Nodes values in the Simulation Parameters ensure that the effective plan area of a node can never fall to zero."