Problem drainage on a very steep street 3

[Boltricity]

I've been asked to look at some drainage problems on an existing fairly long and steep street (10% grade). It was designed with a 2% crown. It is a heavily traveled street with two or three lanes in each direction.

The issue is that during heavy rains there is a lot of water on the pavement which causes vehicle spinouts. Much, if not most, of the water is not getting over to the curb and gutter (and thus the drainage inlets) but instead is “sheet flowing” down the road. This is not surprising when you look at how the pavement is contoured. (A 10% longitudinal slope vs. 2% cross-slope results in contours running almost perpendicular to the road alignment). It's almost as if there was no pavement cross-slope at all.

Does anyone have any experience dealing with something like this, either in the original design or to remediate an existing problem? Obviously in a new design, one could increase the 2% cross-slope some. Too late for that here, unless a complete overlay is done. And since there is an existing raised median, the median would probably have to be torn out and raised.

Someone has suggested installing trench drains in the pavement, but I'm not sure they would be that effective.

Thanks for any suggestions provided.

 

[francesca]

You could do a series of slotted vane drains across the road. Like this.

 

[Kuhuh]

Have you looked into repaving with open graded asphalt? It will allow some of the water to permeate lower and decrease the amount of sheet flow. This may be cosmetically more beneficial for the owner. Just watch out if sediment in runoff is sandy then open graded is well crap.

 

[MiketheEngineer]

How about a sign "Slippery when wet"

 

[Ryb01]

Depending on the flow and design requirements, you may be surprised how many CB's and/or the length of trench drain required to capture the flow. Trying to capture runoff on a slope like this is challenging.

Is there an opportunity to capture some of this runoff prior to flowing down the 10% grade? Are you in a region that experiences a freeze/thaw cycle? How much flow is "alot of water"?

 

[Boltricity]

Hey everybody,

Thanks for the suggestions so far.

Ryb01, I haven't actually calculated the amount of water, but I think it is fairly significant. This is a very long and steep road and there are several curb inlets upstream of the "problem area", but (as I described earlier) most of the pavement flow isn't getting over to the inlets but is just sheeting down the road. So it’s definitely more water than you would want to see on the pavement.

Most of the flow is actual runoff from rain hitting the pavement. Some is probably draining off the median. I don’t think the median runoff is contributing that much to it, as it is landscaped and thus has a much lower runoff coefficient. But it is something. If the median runoff was picked up in the median, it might reduce the overall flow in the street by 10 to 15%. That’s just a gut feeling…no calcs done.

I don’t think we necessarily have to pick up ALL of the street sheet flow, but rather, decrease it significantly. So maybe a combination of 1) keeping the median runoff in the median, and 2) putting in a series of trench drains or vane drains (as suggested by francesca) would be the best solution here.

 

[TGLG]

Slot drains may be your answer.

 

[bimr]

The long term solution is probably cross-slope drainage.

The slot drainage solution may work but maintenance will be an issue as well as the poor capture characteristics of the slot drains. Not sure that it will be practical to keep the drains free of debris.

 

[EngWade]

The trench drains or "vane drains" do not seem like much of a solution to me. You said the road is 5 lanes wide (plus a median), so I'm assuming we are talking about 65' of street width approximately? If that is the case, I'd have to say the best option would be to redo the road. Granted you probably cannot change the 10% longitudinal slope, but could you increase the horizontal cross pitch or reduce the cross pitch length? Meaning, increase the cross pitch to 3%, or crown each travelway (e.g. pitch to the outside curb as well as into the median curb). In my opinion, you're going to have a hard time getting runoff to travel ~30' transversely on a ~10% longitudinal slope.

 

[3feethighandrising]

You could retrofit to a M Road Section /\/\ by removing the existing 40' center section out amd regrading the center lanes section to 2% as a valley (keeping the outermost and inner most existing lanes intact and sloped toward the curb). This would allow you to keep one lane functional in each direction during construction. I'd do catch basins every 150' down the center valley low point with vaned grates. This would more than double the water catching capacity you have now and shorten the longest flow path of sheet flow to a catch basin by 1/2. This will require adding a new storm line running down the center of the valley and connecting somewhere at the bottom.

 

[Ryb01]

I think EngWade brings up some vaild points.

As I hinted toward earlier, this is not going to be a challenging project that will require some "outside of the box" thinking to be able to capture the flow. I think re-doing the road to create a defined cross fall is important. The challege that I forsee, is the "splash over" component of the flow you are going to have to deal with when desiging your inlets. You'll probably find that the inlets (regard less of type) will not be that effective due to the 10% grade. The advantage of using a trench drain is this configuration of inlet will capture quite a bit more flow than a inlet built into the curb. If there is an opportunity to capture a larger portion of the flow on a "flatter grade", it may be worthwhile looking into.

 

[Ryb01]

Another thought....if traction on this road is a concern...is there an opportunity to repave with concrete? A bit rougher finish than usual? Maybe consult with your local municipality and/or DOT.....

 

[cvg]

reconstruction will be very expensive. Adding sufficient slotted drains at close enough spacing to eliminate the problem will also be very expensive.

spinouts are caused by a combination of a)too much water on the road causing hydroplaning and b)high vehicle speeds
If you can't economically reduce the amount of water than reduce the signed speed and slow down the traffic.

 

[Boltricity]

Hey all,

I appreciate the ideas and comments everyone is posting.

I think I should clarify what the current street cross section looks like. This road is designed with three lanes and a bike lane (41’ curb to curb) in the south (uphill) direction, and two lanes and a bike lane (29’ curb to curb) in the north (downhill) direction. There is a 12’ wide median, but it is pitched toward the downhill side. (The road centerline and crown is set coincident with the west curb line of the median.) The street has a 50 mph design speed and is signed for 50 mph.

The problem area is in the north (downhill) direction just into a 1400’ radius curve (left) after a very long downhill tangent section.

As I see it, the problem is a combination of:
1. Too much water on the pavement (during heavy rain) for the reasons described earlier.
2. Many drivers likely exceeding the speed limit due to the long, steep downhill run.
3. Drivers entering a curve to the left when they are already on pavement cross-sloped to the right.

Ryb01, your idea to create a rougher road surface I think is a good one. Obviously there is a coefficient of friction factor here that is lacking and contributing to the problem. I think reducing the speed limit to 45 mph and adding more signs would help. I don’t think there are enough speed limit signs along that stretch of road as it is.

But I still think the amount of water on the pavement needs to be reduced. I don’t think retrofitting to the “M” road section would be practical in this case, as it would require too much reconstruction of the road. But I think the easterly median curb could be rebuilt at a higher elevation and the northerly (downhill) lanes overlaid to get to a 3% to 4% cross-fall.

kpanzer, the open graded asphalt is not a bad idea. I think sediment can be controlled, as this is primarily pavement runoff. This could possibly be incorporated into the overlay.

Whether or not to add slotted vane drains, would need to be evaluated based on the newly calculated amount of runoff on the pavement, and the capacity of the drains vs. cost.

 

[beej67]

Obviously the first thing to check is to make extra sure that there's no offsite run-on slopping into your road somewhere. That can kill you, and it's the sort of thing that road designers sometimes miss if they don't have good survey to design to when they do their vertical alignments. I know back when I was doing road design, I was asked to design roads on some very sketchy topo.

After that, I definitely think adding a top layer of "open graded friction course" is a real winner. That stuff works wonders in this area of the country. Really amazing pavement. Add your crown while you're repaving.

Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[EngWade]

How does your road perform in the winter? What type of inlets do you have? Curb inlets?

Could you build up the left curb of the downhill median enough to pitch it to the west, into the uphill travel way? That would help eliminate some runoff from the median reaching the problematic (downhill) travelway. Doing this, could also help you achieve a more aggressive cross slope on the downhill travelway. To me, this is the only real option that would provide much benefit based on the conversation so far.

I'd be skeptical of the benefits you'd actually realize from surface roughening on a 50mph design speed, 10% longitudinal (downhill) grade, with a reversed super elevation.... in wet weather.

One last question, whose issue is this? State DOT I assume? Unless you work for DOT (obviously), you might check with them on any recommendations.

 

[CPENG78]

Here is my 2 cents,
From reading the posts, it would seem to me that doing something other than repaving the road at a higher cross slope may be just, "hiding" the original lack of thought that went into the original design (No disrespect meant to the posted ideas). Engwade brings up great points, as it is obvious that the road should have a higher cross slope in this situation. I was involved in a project that was fronting a road that If I recall correctly included a 5% longitudinal slope and the cross slope varied from 0% at the crown to almost 5% by the time you reached the gutter pan(basically a parabola across the road with). A 10% cross slope in your situation makes it even more challenging. I'm curious as to the obligations of the client in relation to what fronts this road. Like beej67 and Ryb01 mentioned, I would also consider accounting for all water that makes it to the road. Certainly, it would be alot of water produced by the road, but stopping water at the top of the hill and making sure that no water from property along the frontage are not contributing to the problem is certainly worth looking into. In addition to inreasing the cross slope, I would also look into replacing or installing additional inlets with longer curb openning and/or grate opennings to prevent water from shooting past the inlets. Of cource, I do understand that sometimes we are tied by what is economically feasible for the project at the time of design/construction.

 

[TerryScan]

The capture characteristics of a slotted drain can be actually quite good. Also, clogging issues are probably a lot lower on such a slope compared to installations on low grades in a sag where debris will accumulate.

 

[Boltricity]

I want to thank everyone so far for weighing in with their questions and suggestions regarding this problem. There are many solutions being considered, and the final one may end up being a combination of two or more of these.

I thought I'd update everyone with some more clarifications and where I'm at with this project.

1. Major reconstruction of the road cross-section (raising the median and overlaying to increase cross-fall, or creating an "M" cross section) is probably not feasible at this time. I don't think that increasing the cross-fall (from 2% to something like 4%) would be a good idea after all, as part of the problem is drivers entering a curve to the left with the pavement cross-falled to the right. That could potentially just make the spin-out problem worse.

2. Reducing the posted speed limit would probably not be effective, as many drivers would likely exceed it anyway on the long downhill run. However, see #3.

3. Increased speed limit signage and also new warning signs for drivers to watch downhill speeds in wet weather and/or approaching the curve is definitely a must, regardless of other mitigation measures taken. There is insufficient signage as it is. Also, a guardrail and chevron signs are being considered for the initial part of the horizontal curve.

4. Another possibility is rumble strips across the road, approaching the accident area, to encourage drivers to slow down. I've seen these used locally, even on freeways where the end of freeway is approaching (example, at a border or other checkpoint). Anyone have any thoughts on these?

5. The vane drains are being seriously considered. It would be good to see some design standards on these. Does anyone know if any local or state agency is using these, and has standard drawings or design information for them? Related to that, is there a standard for determining the drainage capacity of the vane drains? I've done some preliminary calcs and I think we need to capture somewhere between 0.8 and 1.0 cfs over roughly 120 feet (longitudinal length) of roadway. A partial capture could also be useful.

6. Sawing grooves at an angle in the pavement has also been considered. This would help get drainage off the road surface quicker, reduce the thickness of the water "film" on the pavement surface, and also provide some traction on the roadway. Does anyone have any experience or thoughts regarding this?

7. Additonal mitigation being considered is doing a chip seal on the asphalt roadway to increase traction. This could be done in addition to the other mitigation measures.

Again, thanks to everyone for chipping in with your thoughts and suggestions.

By the way, when I initially posted this issue it was my first time on this web site, and I want to say this is an incredible site with the ability to share tips, ideas, problems and solutions with other engineers. Whoever came up with this site deserves some kudos.

 

[swazimatt]

You could consider placing the rumble strips at an angle instead of sawing the road. These would obviously need to be placed before the curve as they tend to reduce braking efficiency and tyre traction - don't want them to increase the risk of accidents

I have seen speed humps used in townships to direct stormwater into drains, but not at 50mph!