Three-Dimensional and Uniform UV Curing Using Integrating Sphere Technology

By Thomas Efsen, managing director, Efsen UV & EB Technology

An integrating sphere uses multiple diffuse reflections of light incident on the sphere inner surface to homogenize the light over the entire surface (Figure 1). This results in uniform irradiance at all points on the interior surfaces of the integrating sphere.

This has been applied to UV curing by recognizing that uniform irradiance on the interior surface of the sphere creates uniform irradiance at all points within the volume of the sphere. Thus, an object placed inside an integrating sphere will be exposed to a relatively uniform level of radiation on all its surfaces facing the sphere wall (within certain limits of approximation).


The integrating sphere for UV curing applications is a hollow chamber with diffusely reflecting inner surfaces. Light from a UV source is coupled to the chamber via one or more ports in the chamber wall, and objects to be cured are placed inside. Multiple diffuse reflections from the chamber walls lead to uniform irradiance on the chamber walls. Uniformity level is close – but not equal to – an integrating sphere system, but achievable within +-5% due to the rectangular geometry of the used chambers.

The result is a spatial integration of the light source geometric information, completely decoupling the light field inside the chamber from the lamp geometry. All points within the interior volume of the integrating sphere then have the same irradiance. This has several specific benefits for UV curing:

  • Uniformity of irradiance. All surfaces to be cured receive the same irradiance.
  • 3D exposure. Irradiance is the same on all surfaces, not just those facing the lamp. Objects do not need to be rotated or moved.
  • Radiant exposure is decoupled from light source. The lamp can be smaller than the treated object. Multiple parts can be exposed using a single lamp. No part or lamp motion is required.
  • Tighter process control. The irradiance is known at all points inside the integrating sphere, so the actual part exposure can be monitored in real time.

Applications to UV Curing

This technology is successfully being used to cure coatings and adhesives. Following are some examples.

Catheter/guidewire coating. A customer now cures with an integrating sphere curing chamber using two electrodeless UV lamps to cure 10 parts at once (each ~1200 mm long) within 30 seconds with zero cure-related scrap.

Catheter coating. An existing curing system used a traditional approach with rotating catheters individually in front of a bank of arc lamps. After installing an integrating sphere-based curing system, their throughput increased more than an order of magnitude, curing multiple catheters in a single batch with no rotation and higher finished product quality.

Adhesive curing. A customer’s product contained several UV-cured adhesive joints. These were cured one joint at a time by placing the uncured parts on a conveyor, requiring four passes as the part assembly progressed. By converting to the integrating sphere technology, the customer now cures all four adhesive joints on two separate parts – eight adhesive joints total – at one time, with a single 30-second exposure.

Test objectives and setup

Our testing focused on these questions:

  • Uniformity of irradiance: Is it really uniform?
  • Irradiance stability: Does the irradiance vary with time and from exposure to exposure?
  • Energy control: How repeatable is the energy, in J/cm2, on exposed parts?
  • Thermal effects: Do the parts experience more or less heat than with conventional direct irradiation?


Chamber. The chamber used was a VelaCure™ system (Figure 2) from Vela Technologies. It is a laboratory-use curing system, designed for six catheters or guidewires. The main elements of the system are the chamber, shutter, controller and UV lamp. Parts are loaded manually on this particular unit.

Parts may be suspended from a manually loaded rack, as shown in Figure 3 (urinary catheters shown). The rack is then placed over an opening in the chamber top, and a safety door closes over the rack, preventing UV escape from the chamber.

Pressing the RUN button on the touchscreen control unit causes the shutter to open, letting UV into the chamber and beginning the cure cycle. (The lamp is normally always on.) Once the target energy (in J/cm2) is reached, the shutter is closed and parts are removed.

The chamber also has a holder for attaching an EIT Power Puck radiometer to measure the chamber irradiance.

Lamp. A Heraeus Noblelight F300 lamp system with an H-bulb and standard reflector was used as the UV source. This is mounted on the chamber with the shutter between the two, with the chamber’s control unit used to control all lamp functions.

Diagnostics. All radiometric measurements were performed with either a Power Puck II Profiler Mid-Range or a MicroCure 2W radiometer from EIT. Only UVA values were recorded (defined as 320 to 390 nm by EIT).

Two types of UV-sensitive color-changing strips from Con-Trol-Cure, Inc., were used to visualize uniformity on the chamber walls and object surfaces. Both are paper strips that change color in proportion to incident irradiance, providing a visual reference for relative exposure. The first type has five exposure zones, each zone having decreasing sensitivity to incident UV (“5-zone strip”). This type of strip makes it easier to compare color changes, as there are five different shades of color that must match, so as to evaluate each energy level. The second type has only one color-changing zone (“single-zone strip”). This is used to evaluate uniformity, as change in shade level indicates different exposures. It should be noted that pictures taken of these strips do not fully depict viewing the strips side by side, as lighting, shadows and camera settings do not offer a result equal to visual inspection.


One of the promises of this technology is uniform, 3D exposure of multiple objects simultaneously.

Testing of uniformity was divided into two efforts:

  1. Uniformity on chamber walls. If the chamber is truly acting as an integrating sphere, then the irradiance on each wall should be the same.
  2. Uniformity on multiple parts. Finally, we suspended various parts inside the chamber and exposed them together. This is how parts are normally processed through the chambers in production and represents the most difficult situation to cure.

Uniformity on chamber walls. Five-zone color strips were placed on each chamber wall except the rear and exposed together (Figure 4). The radiometer was used to record the absolute irradiance. This should provide a relative measure of irradiance uniformity on the chamber walls.

Figure 5 shows the five-zone test strips that were placed on five of the chamber’s six walls and exposed, along with the radiometer value during that exposure. The radiometer was on the right wall. To get the irradiance within the strip’s sensitivity range, the shutter was opened 20%, meaning the chamber irradiance was approximately 20% of its maximum value. All strips were exposed together for 5 seconds.

There is no detectible difference in color change of the strips when comparing correlated zones. All of the color changes correspond to 150 to 200 mJ/cm2 according to the comparison chart supplied with the strips. The radiometer reading is 101 mJ/cm2, quite a bit less than the color chart indicates. The difference could be due to the larger field of view and of the flat test strips.

Uniformity on multiple parts. For this test, we suspended three different parts inside the chamber using the parts rack. Figure 6 shows the setup. The parts used were a urinary catheter (clear), a black plastic car dashboard part and an aluminum pipe.

The results are remarkable. Figure 7 shows the exposed strips. In each case, the exposed colors correlate with 70 to 100 mJ/cm2 – except the last aluminum pipe and last dashboard strip, which correlate with 50 to 75 mJ/cm2. The former was on the inner surface of the pipe; the latter was on the inside (concave) surface of the dashboard part. Note that the “Vela wall” strip is almost indistinguishable from the others.


Uniformity conclusions. Based on these tests, we find that:

  1. Chamber walls have uniform irradiance.
  2. Multiple parts placed inside the chamber experience very high uniformity of irradiation.
  3. Heavily shadowed regions, such as the aluminum pipe inner surface or dashboard concave surface, still received sufficient irradiance for UV curing applications, with a 25 to 50 percent reduction in irradiance, all depending on geometry.

Irradiance stability

Having shown the chamber irradiance is uniform, we wanted to test the stability with time. Because the chamber is simply an assembly of diffuse reflecting walls, one would expect the time-dependent irradiance to follow that of the lamp used.

To investigate this, we recorded the chamber irradiance using the profiling radiometer during repeated exposures. This profiling radiometer records the irradiance vs. time at a sampling rate of 1024 samples/sec.

Repeated exposures. We performed a set of tests with the controller set in “dose mode” to evaluate stability and repeatability of irradiance during exposures typically used for production. In this case, we ran the system in dose mode. Each exposure lasts ~30 seconds, and the time between exposures was ~30 seconds as well. The lamp was on continuously, and exposures were controlled by opening and closing the shutter.

Results are shown in Figure 8, with values of each exposure given in Table 1. Figure 9 shows the average irradiance of each exposure. Once again, the irradiance is stable during each exposure. Overall, all 10 exposures averaged 139.4 ± 0.359 mW/cm2 (±0.26%). We believe the irradiance variation within each exposure is due to lamp output variation but did not attempt to verify this.

Stability conclusions. Chamber irradiance is repeatable, within 1 mW/cm2 in the condition tested.

Dose control

One of the more intriguing features of the system used is what is called “dose control.” Recall that a part exposed to UV starting at time t1 and ending at time t2 receives energy (dose) given by the irradiance on the part integrated over the exposure time. However, as shown above, irradiance on the part is the same as irradiance on the chamber wall, due to chamber uniformity. Therefore, the energy received by the part is:

So, by measuring the irradiance at any point on the chamber wall during an exposure, the energy delivered to all of the parts inside the chamber can be calculated in real time. Once the calculated energy reaches a user-set target value, the controller automatically stops the exposure.

[Note: Dose is sometimes referred to as “fluence” or “energy.” All are typically used to mean energy received per unit surface area. “Exposure” in this paper refers to irradiating a part, not the energy (dose) received.]

The result is that variations in irradiance over time and from exposure to exposure can be automatically compensated using feedback control to give the required dose in all applications and on all parts. A further benefit is that the same chamber configuration can be used for a variety of parts.

Repeatability. We evaluated the chamber performance when in dose control mode using the “repeated exposure” tests above and calculating the energy received for each exposure. The target was 4.1 J/cm2. The time-dependent measurement is shown in Figure 10 (along with irradiance for clarity). Total energy (dose) per exposure is shown in Figure 11, averaging 4.12 ± 0.014 J/cm2 (±0.341%) over all exposures.

Effect of parts on energy control. We wanted to know if delivered energy was consistent as the number of objects inside the chamber varied. To do this, we kept the target constant while exposing from 0 to 6 identical urinary catheters and measured the energy (integral of irradiance) using the profiling radiometer.

Results are shown in Figure 12. The irradiance changes as the number of parts inside increases, as discussed above: Each added part causes the irradiance to drop due to increased light absorption. To keep the energy delivered constant as parts are added and the irradiance drops, the VelaCure system automatically increases the exposure time. The net result is a constant energy delivered to all parts inside the chamber.

Control conclusions. The system tested exhibited very good control of energy delivered to exposed parts.

Thermal effects

Parts heat up during exposure due to absorbed UV, visible and infrared radiation. Uniformly irradiated parts should, therefore, heat uniformly.

We tested this by exposing a complex-shaped black plastic part using both the chamber and a conventional direct-exposure method – a 120 W/cm electrodeless UV lamp mounted over a conveyor belt. For the direct-exposure method, we measured the energy density by stopping a conveyor when the PowerPuck II filter was directly under the UV lamp – but at a distance exceeding 25 mm from the focus point. We had 1300 mW/cm2 for about 7 seconds, exposing the unit to 9,1 J/cm2. We then placed the same part in the chamber and exposed it via Dose Control mode to 9,1 J/cm2 (duration about 150 seconds). The results are shown in Figure 13. The directly irradiated part partially melted during the exposure. The part exposed in the chamber showed no damage with the given energy density.


We evaluated an integrating-sphere based UV curing chamber. Our performance criteria were irradiance uniformity, repeatability of irradiance, consistency of delivered energy density and reduction of localized heating. In all cases, we found this technology performs exceptionally when applied to 3D UV curing.

For the last 10 years, Thomas Efsen has been owner of EFSEN. EFSEN has been developing optimal UV & EB solutions for the medical device industry, among others, since 1986. For more information, call +45 4565 0260, email: or visit