This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DLPDLCR4710EVM-G2: Defects in the continuity of the light intensity

Part Number: DLPDLCR4710EVM-G2
Other Parts Discussed in Thread: DLP4710EVM-LC,

Hi,

We are currently using the DLPDLCR4710EVM-G2 and DLP4710EVM-LC for some experiments which need to control the intensity of the light. We use the optical power and energy meter console from THORLABS to examine the optical power, and found that both products have some defects with the continuity of the light intensity. We have confirmed the defects also by the luxmeter and the Basler camera.

The DLPDLC GUI setting is as following:

We give the DLP a homo intensity picture which is a 8-bit one to get the optical power. Here's the DLPDLCR4710EVM-G2 data. The first column refers to the 8-bit blue channel. The second column refers to the radiometric we get from the optical meter.

TI_E2E.xlsx

TI_E2E.txt
"Gray Scale"	"Radiometric(W/m^2,470nm,raw)"
0	0.001
1	0.001
2	0.001
3	0.001
4	0.001
5	0.002
6	0.002
7	0.002
8	0.002
9	0.002
10	0.002
11	0.002
12	0.003
13	0.003
14	0.003
15	0.003
16	0.004
17	0.004
18	0.004
19	0.005
20	0.005
21	0.005
22	0.006
23	0.006
24	0.007
25	0.007
26	0.008
27	0.008
28	0.009
29	0.009
30	0.01
31	0.011
32	0.011
33	0.012
34	0.013
35	0.013
36	0.014
37	0.014
38	0.015
39	0.016
40	0.017
41	0.018
42	0.018
43	0.019
44	0.021
45	0.022
46	0.023
47	0.024
48	0.025
49	0.026
50	0.027
51	0.028
52	0.029
53	0.03
54	0.032
55	0.033
56	0.034
57	0.035
58	0.036
59	0.038
60	0.039
61	0.041
62	0.042
63	0.044
64	0.045
65	0.047
66	0.048
67	0.05
68	0.051
69	0.053
70	0.055
71	0.056
72	0.058
73	0.059
74	0.061
75	0.063
76	0.065
77	0.066
78	0.068
79	0.07
80	0.072
81	0.074
82	0.076
83	0.078
84	0.08
85	0.082
86	0.084
87	0.086
88	0.089
89	0.091
90	0.093
91	0.096
92	0.098
93	0.1
94	0.103
95	0.105
96	0.107
97	0.11
98	0.112
99	0.114
100	0.117
101	0.119
102	0.122
103	0.124
104	0.127
105	0.13
106	0.132
107	0.136
108	0.138
109	0.141
110	0.144
111	0.147
112	0.15
113	0.153
114	0.155
115	0.156
116	0.159
117	0.162
118	0.165
119	0.168
120	0.171
121	0.174
122	0.177
123	0.181
124	0.184
125	0.187
126	0.191
127	0.194
128	0.198
129	0.201
130	0.205
131	0.208
132	0.212
133	0.216
134	0.219
135	0.223
136	0.226
137	0.23
138	0.234
139	0.238
140	0.242
141	0.246
142	0.249
143	0.253
144	0.257
145	0.26
146	0.264
147	0.267
148	0.271
149	0.275
150	0.279
151	0.284
152	0.288
153	0.292
154	0.296
155	0.302
156	0.307
157	0.309
158	0.313
159	0.317
160	0.321
161	0.331
162	0.337
163	0.335
164	0.34
165	0.344
166	0.349
167	0.353
168	0.363
169	0.366
170	0.367
171	0.371
172	0.376
173	0.381
174	0.385
175	0.39
176	0.395
177	0.4
178	0.405
179	0.411
180	0.415
181	0.42
182	0.425
183	0.43
184	0.435
185	0.44
186	0.445
187	0.45
188	0.442
189	0.442
190	0.45
191	0.454
192	0.458
193	0.463
194	0.468
195	0.475
196	0.48
197	0.486
198	0.491
199	0.5
200	0.507
201	0.512
202	0.517
203	0.522
204	0.528
205	0.534
206	0.539
207	0.545
208	0.551
209	0.558
210	0.563
211	0.569
212	0.575
213	0.58
214	0.586
215	0.592
216	0.6
217	0.605
218	0.609
219	0.615
220	0.621
221	0.627
222	0.633
223	0.639
224	0.647
225	0.649
226	0.647
227	0.65
228	0.654
229	0.66
230	0.668
231	0.676
232	0.682
233	0.691
234	0.702
235	0.708
236	0.714
237	0.721
238	0.727
239	0.733
240	0.74
241	0.746
242	0.752
243	0.756
244	0.762
245	0.769
246	0.775
247	0.783
248	0.789
249	0.794
250	0.799
251	0.805
252	0.812
253	0.819
254	0.826
255	0.826

When the input gray scale meets about 185, there's a defect.

The other product DLP4710EVM-LC has a defect at about 177.

We are confused that whether we have a mistake in the DLP setting or there's other possible problems during the experiments.

Any hints would be appreciated.

Best Regards!

  • Hi,

    Welcome to E2E thank you for your interest!

    have you tested vertical and horizontal ramp test patterns? is there any discontinuity seen?

    Regards,

    Arthur

  • Thanks a lot for your reply.

    Yep. We offer the DLP some ramp pictures, and the discontinuity is obvious from the camera. Here's the picture we use HDMI to display.

    Here's another picture took by the basler camera, when we use the DLPDLP GUI to display the color ramp patterns

    The darkness at the edges is due to the ramp pattern and the aperture size of the lens.

    Best Regards.

  • Hi,

    Thank you for this information. I will get back to you with any information I find early next week

    Regards,

    Arthur

  • Hi,

    I reproduced this issue. I looked into similar documented issues and it seems that this non linearity is due to the LED illumination and the data being displayed by DMD not being aligned perfectly. 

    the LEDs take a small time to reach full brightness when they are turned on and a small time to reach no light when turned off. if parts of these transitions appear when some micromirrors are modulated "on" you will end up with some non linearities

    Can you tell me more about your application? are you using a single LED for illumination? If you must use all 3 LEDs can you tolerate less brightness?

    Regards,

    Arthur

  • Hi,

    Thank you for your reply. 

    We use the DLP to establish a diy microscope in order to observe the motion and the formation pattern of the algae under the specific light environment. The algae are very sensitive to 470nm blue light. This defect affects the motion which may make the results unconvincing. 
    Usually we turn on the blue channel and disable the red and green LED. 

    Best regards. 

  • Hello Li Yuan,

    Based on the GUI screenshot above it appears you are running the PMIC at the minimum (or near minimum) current level. I would advise that the PMIC may suffer from poorer linearity performance at the bottom 15% of the full scale range.

    The current output given is calculated based off a combination of the DAC value (0-1023) of the current setting register in the DLPC34xx and the Rlim resistor value which is used as a reference to drive the LED current. If you want to run the system with lower current without compromising on image quality you may choose to increase the Rlim resistor value on the board. Note that this comes at the cost of also reducing the current output at full scale (1023). 

    I hope this helps.

    Regards,

    Philippe Dollo