Skylights Vertical vs Splayed By Lisa Bornemann Brad

Skylights Vertical vs. Splayed By: Lisa Bornemann & Brad Koehler

Goals For our final project, we studied the skylight system found in the University Support Building, located across Park Avenue. There are large skylights through the main corridor of the support building. We chose to study the changes in illuminance levels while varying the parameters of the skylight. In this presentation, we took actual measurements with an illuminance meter and we ran Radiance for the same situation. These numbers turned out to be quite different in magnitude, but the patterns of higher or lower illuminance values were the same. These values may differ because the dimensions of the real skylight well, the height of the glass pyramid, and the width of the mullions were eye balled. Also, while we were taking the measurements, the sky conditions were relatively clear, but there were clouds in the sky that would block the sun at times. We waited patiently to take the measurements when the sun was not being shaded by clouds and the area of the sky above the skylight was relatively clear. Still, with some puffy clouds in the sky, they may have affected the results as compared to the description of what clear sky conditions is on Radiance which is the condition we used to run our tests. Throughout the presentation we have shown our different modeled situations and the illuminance measurements obtained through running Radiance. We will compare these measurements and explain how the different parameters change the output data. The actual skylight has a 10 foot well with a 9 ft. by 9 ft. opening. The glass is in a 3 ft. high pyramid shape with 4 in. mullions. We have chosen to discuss four modified situations: a 5 foot well, no well, a 5 foot splayed and 5 foot vertical well, and a 5 foot splayed with no vertical well. For a corridor, only 5 footcandles is required according to the IES handbook. The average illuminance with the current skylight is about 250 footcandles. So obviously the requirement is met. In this presentation, we are not going to determine which skylight is the best, but we are going to explain how the illuminance levels change with the different situations. This information can then be used to determine which skylight will be the best design for particular situations requiring different illuminance levels.

Information Used Ø The transmittance of the glazing was calculated by taking the illuminance outside and the luminance inside aimed at the skylight glass. Lemitted= 805 cd/m 2 Eincident = 4900 lux Transmittance = Π*Lemitted / Eincident = 0. 52 *Note: Transmittance used in Radiance is 0. 534 Ø Ø Ø Sky conditions for all situations were considered clear. The date the measurements were made and calculations were ran for April 18 th. All actual measurements were made between 2 pm and 3 pm, so in Radiance, the time used was 2: 30 pm. All illuminance measurements shown throughout the presentation are in footcandles. Throughout the presentation, four corner points will be discussed. Corner 1 through 4 in the counter clockwise direction. The color key below will help make them easier to reference: corner 1 corner 2 corner 3 corner 4 Ø Ø

Original Design ~ 10 ft. Well Actual illuminance levels obtained by taking measurements with an illuminance meter at the site: 230 243 262 274 283 293 279 254 239 536 245 288 284 302 298 291 289 274 464 451 260 278 289 303 313 302 305 295 275 453 447 439 259 279 284 293 306 300 289 274 486 479 472 464 244 255 269 264 295 298 290 279 264 496 489 456 449 401 234 245 258 264 269 274 270 263 248 357 354 492 441 429 416 229 236 247 253 258 265 252 243 235 366 382 402 396 434 420 373 221 229 240 248 251 257 243 238 229 374 399 395 437 395 385 363 216 222 233 241 243 249 238 231 220 489 512 455 435 455 497 502 527 516 532 461 467 444 473 468 548 363 430 449 435 485 517 538 383 443 452 420 427 456 407 415 424 431 453 415 401 412 440 393 400 359 342 321 344 322

5 ft. Well In this model, the illuminance striking the floor was higher than the illuminance on the floor of the 10 ft. well design. We feel that because this well is smaller than the previous, there are fewer bounces of reflected light. The result is a higher illuminance level across the floor because less light is absorbed by the well walls. 406 451 566 567 536 459 442 396 398 433 392 504 442 434 451 422 388 399 464 419 471 462 452 470 466 430 386 448 453 443 482 487 476 442 392 375 465 511 502 491 480 463 410 392 567 548 533 592 544 500 482 463 475 605 592 633 613 539 494 532 513 480 631 671 651 632 554 495 484 457 429 691 690 667 645 566 550 568 535 497 Also, you will notice that corner 2 has the highest illuminance of all of the corners while in the 10 ft. well, the diagonal corner (corner 4) has the highest illuminance. We feel this occurs because in the shorter well, more light directly hits the floor opposite the sun. Corner 2 is getting directly hit by the sun’s rays.

No Well In this model, the illuminance striking the floor was higher than the illuminance on the floor of the 5 ft. well design. We feel again that because there is no well, there is no reflected light. The result is a higher illuminance level across the floor because no light is absorbed by the well walls. The pattern formed on the illuminance grid is the same as the 5 ft with the highest illuminance in corner 2. 400 402 404 394 382 372 362 340 319 457 454 449 442 427 419 399 376 350 511 510 501 489 470 463 438 412 378 580 571 554 537 524 499 476 444 408 644 608 595 581 567 538 505 475 435 691 664 652 645 620 583 549 508 459 754 747 716 689 663 629 580 525 476 817 805 758 733 711 671 598 548 498 890 823 797 786 741 658 619 570 513

Splayed w/ 5 ft. Well In this model, the illuminance striking the floor was higher than the illuminance on the floor of both the 5 ft and 10 ft wells. The splayed parts of the well will reflect light downward with less bounces. Also, a splayed well yields a larger source of light. We feel these are both reasons that the illuminance is higher. 445 450 458 455 447 423 407 377 366 474 481 482 473 469 446 403 392 381 508 511 502 499 492 465 447 462 442 541 532 529 514 515 502 496 476 459 565 559 547 534 530 529 509 489 474 584 578 566 552 572 542 523 502 484 626 606 659 629 590 561 539 516 494 685 679 677 647 617 584 550 521 497 729 725 695 665 637 586 557 528 503 The pattern formed on the illuminance grid is the same as the 5 ft and no well with the highest illuminance in corner 2.

Splayed w/out Well In this model, the illuminance striking the floor was higher than the illuminance on the floor of the splayed with a 5 ft well. Less bounces occur because there is no vertical well. Also, a splayed well yields a larger source of light. We feel these are both reasons that the illuminance is higher. 486 487 492 488 479 501 478 453 426 532 524 527 553 532 524 509 483 458 583 601 617 597 576 557 542 514 473 679 663 653 643 621 595 559 530 496 739 723 707 692 656 617 585 559 520 804 770 753 740 702 662 609 576 542 880 847 824 789 750 737 639 603 559 978 903 848 799 834 777 694 624 579 1092 973 918 934 879 820 691 645 598 The pattern formed on the illuminance grid is the same as that of the splayed with a 5 ft well with the highest illuminance in corner 2.

Radiance Rendering Vertical Well

Radiance Rendering Splayed w/ 5 ft Well

Conclusion Ø A shorter vertical well will yield a higher illuminance level on the floor due to less light being absorbed by the well walls. Ø A splayed well will yield higher illuminance levels than a vertical well of similar proportional. Ø A splayed well becomes a larger light source and will therefore spread more light over a larger area, thus increasing the illuminance directly below the well.