Michael,

 

This seemed like a great topic for the roundabout listserve

 

I would like to have better insight to the issues – and appreciate your explanation as to why you prefer radial designs. I agree with you concerning single lane designs, but haven’t seen multilane designs that support your premise, so I am hoping the community will help us both clarify the issue.

 

I have great respect for your significant contributions to the profession overall, and respectfully disagree with your approach to multilane roundabouts. I hope we can both learn something through the listserve. I have no desire to become a roundabout designer, but do believe I may something to contribute in being able to identify specific problems with a design, and that the iterative process is how great designs are achieved.

 

Here is my post:

 

> Michael,

>

> I agree that your statement "that exit speeds at radial design

> roundabouts ... is overall well below the exit speed at Offset left

> roundabouts" is correct. Your conclusion that "because it reduces exit

> speed and thereby makes it safer for pedestrians on the crosswalk of

> the exit lane," seems to me an oversimplification of the thinking

> process, especially at multilane roundabouts.

>

> There may be specific locations where that is an appropriate solution,

> but I don't agree with your premise. From what I have seen, having

> good deflection prior to the roundabout and smooth exits are critical

> for the best multilane designs. Of all the examples I have seen at the

> National Conferences, I don't remember seeing a multilane radial

> design roundabout that was "the best" solution. I've seen radial

> multilane roundabouts that were constructed

> - but they ALL seemed to me to be cookie cutter, rather than designed

> for the location.

>

> Can anyone provide examples (and justification) for good radial design

> multilane roundabouts?

> 

 

Frank

 

Over the past 37 years I have designed many multilane roundabouts starting with the first roundabout, a three lane roundabout, in Australia. My guess is that I have designed more than 200 two and three lane roundabouts in Australia, US, and Bahamas. Every single one of them was specifically designed for the location. Every one was different. Everyone that I have I had post construction crash data have had very low crash rates (a few per year) and very few injury crashes at all. 

 

However there are many roundabouts designed by others that have had a large increase in crashes, 400 and 700 percent after the roundabouts replaced signalized intersection. Radial design does not mean good design. Radial design is the best only when the design is of high quality  with absolute attention to speed control. 

 

The Venice roundabout is a good example of radial that uses the same principles that I use on all of my roundabouts. And the first set of crash data post construction had only a couple of crashes, a rear end and someone ran into the departure curb. 

 

If you like, I will put together a collage of multilane roundabouts that I have designed with some photos of some badly designed radial roundabouts designed by others that had large increases in crashes Just give me a few days. 

 

I can also give you some references to call if you wish to check on my designs.

 

 

Posted on roundabout Listser:

 

 I agree with you Mark, but if you two roundabouts using the same central island diameters, same ICD, same number of lanes then the exit speed at the departure crosswalk will be higher at an offset left roundabout  rather than a radial roundabout. Reason, at an offset roundabout drivers can accelerate at 6.9 ft/sec/sec (NCHRP 572?) from the high point of the central island to the departure crosswalk. At a radial roundabout a driver cannot accelerate until they reach the end of the exit radius about at the depurate crosswalk. NCHPR 572 shows that roundabouts with exit radius control have lower exit speeds than roundabouts with tangential exits upwards of 10  mph an hour. 

 

As speeds increase driver yield rates drop precipitously, UNC Study.

 

My field observations confirm this scenario. 

 

 

Michael Wallwork, PE
Alternate Street Design, P.A.
904 710-2150

 

From: Roundabout Research [mailto:ROUNDABOUTS@LISTSERV.KSU.EDU] On

Behalf Of

Michael Wallwork

Sent: Tuesday, June 04, 2013 10:01 PM

To: ROUNDABOUTS@LISTSERV.KSU.EDU

Subject: Re: Ped & Bike safety at roundabouts

 

 

Speed control requires a designer to fully understand roundabout design, speed control design vehicles requirements. NCHRP 272 shows that exit

speeds at radial design roundabouts, roundabouts where exit speed is controlled

by the exit radius is overall well below the exit speed at Offset left roundabouts.

 

Two studies one by UNC and one done for the Minnesota DOFT show that yield

rates to pedestrians at the departure crosswalks decreases more so than on

the approaches due to vehicles speed, probably because vehicles are

accelerating on the exit rather than slowing on the approaches, Also, that

it only takes a few extra miles per hour to cut Yield rates from 83 to 43 percent.

 

Michael Wallwork, PE

Alternate Street Design, P.A.

 

---

"Context" is probably the most important factor between design types, ICD

sizes, and speeds.  I agree a radial design can function well for single

lane roundabouts in Urban conditions... and can function well for peds if

the exits are clearly visible and not too tight.  However, I disagree a

radial design is the best solution for most other situations.  Consider high

speed approaches, multi-lane roundabouts, suburban or rural roundabouts,

etc... radial designs are not the best solution as they create problems for

other aspects of the safety and capacity (peds & vehicles). 

 

On the one hand, there seems to be a focus of discussion of better yield

rates at exits of roundabouts for peds with tighter exits and slower speeds.

On the other hand, tighter exits create other types of vehicular accidents

in the roundabout (i.e. entry path overlap crashes) and can restrict sight

visibility between vehicles and peds.  Slower speeds (too slow) or abrupt

entry designs on entry often create other forms of accidents in the

roundabout .... 

 

A designer shouldn't use only one design type, rather match the design type

to the specific conditions....  Too much change in one area of a design

(such as tight exits for slower veh speeds at ped crossings) may perceivably

solve one issue but creates multiple and significant operational problems in

other areas.  Ultimately, the designer must balance and solve all of the

differing (and often conflicting) variables together.

 

Scott

 

 

 

Scott Ritchie, p.e. (ca)

President, Roundabout Specialist

Roundabouts & Traffic Engineering

Direct: (928) 284-0295

Main: (928) 284-0366

scott@roundabouts.us

www.roundabouts.us <http://www.roundabouts.us/> 

---

Date:    Thu, 6 Jun 2013 23:17:11 +0000

From:    "Dovey, Dan" <Dan.Dovey@KINGCOUNTY.GOV>

Subject: Re: Balance

 

--_000_217B99B8F24A224C92B6FE89DEEC147D214BDE73MAILQDC1kckingc_

Content-Type: text/plain; charset="us-ascii"

Content-Transfer-Encoding: quoted-printable

 

Gene and Scott,

 

I wholeheartedly agree with this as well.  I've seen a few instances locally where I feel Offset Left designs have been taken a bit too far, and unintended consequences have resulted.  In some cases, entries were tight enough at several arterial roundabouts that an appreciable backlash arose from the trucking community - they couldn't negotiate the tight curves with 18-wheel haulers.  Entry speeds were noticeably lower than necessary for all vehicles as well.  To me, the answer was to relax the entry radii and find balance; some offset left, especially at multilane entries, can be a good thing.  But one agency installed outside truck aprons and rolled curbs on all the islands instead.

 

I think it's always better to eliminate a problem if possible than to mitigate subsequent problems as they arise.

 

Dan Dovey, PE

Traffic Control, Modern Roundabouts, and Barrier Systems

King County Road Services Division, Traffic Engineering

W: 206-263-6144   C: 206-423-0973

 

________________________________

From: Roundabout Research [mailto:ROUNDABOUTS@LISTSERV.KSU.EDU] On Behalf O=

f Gene Russell Sr

Sent: Thursday, June 06, 2013 3:47 PM

To: ROUNDABOUTS@LISTSERV.KSU.EDU

Subject: Balance

  

I like Scott's post re balance. We must not forget balance - and purpose.

 

I am one who in favor of roundabout speeds that are reasonably,  appropriately ( you can substitute similar words here) slow. No faster than goes with  other objectives of the roundabout, and I am not sure we yet know all the complexities of speeds, especially in multilane roundabouts. 

 

For example, let's say we could design a roundabout for 5 (FIVE) mph entry and/or exit speed. My feeling is it would be 99.99% safe from fatalities. In addition to many possible adverse or unintended consequences, would it serve the principle purpose of our roads and streets , i.e., "the safe and efficient movement of goods and services" ? That's what I taught for 50 years anyway.

 

Another far out example. We could cut deaths on our interstate system drastically. Really drastically, by a strictly enforced 40mph speed limit. Would that be appropriate for its intended use ? That would be "light years" out of balance.

 

I see all this discussion ( very good, worthwhile and needed discussion !!) as another reason to have a national Roundabout Safety Audit of 400 - 500 roundabouts of many types, including several of offset types serving  varying scenarios.

  

Gene

---

Mark:

 

Yes, size and other geometric elements affect speed at the exit crosswalk--but doesn't the final speed regardless of geometrics need to be kept in mind in design since the primary factor in walker crash rates and injury severity is speed of vehicles?  This primary variable explains in great part, does it not, RIG suggestions of urban roundabouts being built with smaller diameters than rural ones?  Otherwise, why not build large roundabouts in an urban setting?

 

Am not suggesting that all urban roundabouts control speeds down to the

0-10 mph level where you enter "shared space" conditions where virtually all vehicles yield to the walker.  But we need to be cognizant that in some cases that may be desirable depending on the context--and that shared space does appear to meet access to all persons with a handicap if wayfinding is provided and and/or 0-10 mph levels may may also do the same (this latter needs more study).  Poynton, U.K. stands out as a reminder of what is possible, although roundells are the circular treatment at that village center with 26,000 daily entering vehicles.

 

                  Tony

---

Tony, Quick response to an important and large issue:

 

In my experience the issue is not so simple as to suggest radial or non-radial exits are the solutions as you point out and is a significant issue discussed in NCHRP 674 (roundabouts and peds.)

 

Speed control at exit is not solely a function of the exit curbing there

are many factors that are involved including overall size (ICD), shape, and

many other design elements in addition to context and the speed environment

that all play a role.  

 

MTJ