
with you agree. something is. Now..

The Dual-Link DVI Repeater automatically compensates for signal degradation in long cables. Input cable length can be up to 15 m, and the unit can drive long DVI cables on its output to 12 m by boosting the DVI video output.
The DDC channel (for EDID and HDCP) can either be bypassed through the DVI repeater (source “sees” the connected LCD), or emulated, where the EDID is supplied from internal EDID memory in the repeater. When EDID is emulated, HDCP is turned off, forcing the source to send non-content protected video without HDCP. Using the learn button, users can copy and store EDID from any HDMI or DVI LCD into the internal EDID memory of the Dual-Link DVI Repeater.
The unit is configured to buffer single-link DVI or HDMI signals. Using a pushbutton, its mode can be changed to support dual-link TMDS signals.
The unit provides driver free USB 1.1 port connection to a PC. Free Windows™ PC software available on the website, allows reading, saving, manipulating, and writing EDID tables to and from the device. The EMX-DVI modes (Bypass vs. Emulate or Single-Link vs Dual-Link buffer) are also controllable via the USB port.
When configured for Single Link extension, often the unit can operate from the power available through the DVI source signal and no external power supply connection is required. The unit ships with a universal power supply to use if the power through the DVI input is insufficient. The power can also come from the USB connector on a PC.
The Emulated (internal) EDID data can be acquired from any connected LCD (using LEARN function), or a custom EDID file can be uploaded via the USB port. In addition to the above, the unit provides DVI-I bypass for analog RGBHV (or VGA) signals, so it can manage EDID for VGA sources and displays using inexpensive VGA to DVI cables or adapters.
With the included software, you can read and save EDID data from any display.
Approvals — CE, FCC Class A
Environmental —
Operating Temperature: 0–40°C (32–104°F);
Storage Temperature: -40–85°C (-40–185°F);
Humidity: 10–90% non-condensing;
Cooling — Convection
Formats Supported — Video: DVI 1.0, HDMI™ 1.4, HDCP 1.0, Analog RGBHV
MTBF — 90,000 hours
Power — 5V DC, 1A 90-264 VAC, 47-63 Hz
USB Version — 1.1 full-speed
Dimensions — 31.75H x 71.76W x 68.83D mm (1.25" H x 2.7" W x 2.8" D)
Weight — 0.5 kg (1 lb.)
• (1) Dual-Link DVI Repeater
• (1) 5 volt, 1 amp power supply with power cord
• (1) user manual
• (1) CD with software
CONFIDENTIAL. Limited circulation. For review only. Active magnetic bearing: a new step for Model-free control Jérôme De Miras, Samer Riachy, Cédric Join, Michel Fliess and Stéphane Bonnet Abstract— Model-free control is applied to the stabilization of an active magnetic bearing, which is quite an important Regulator Regulator industrial device. Experimental results are compared to those obtained via other control techniques, showing at least on-par Force Force performance with this very straightforward approach. I. I NTRODUCTION Fig. 1. An active magnetic bearing axis Most uses of active magnetic bearings (AMB) are found in industrial applications. In particular, they find their way into high-speed rotating equipment such as turbines, machine tools, vacuum pumps or compressors. Another significant along two degrees of freedom is obtained by combining two use is flywheel-based energy-storage devices, in applications control axes. to completly maintain a shaft in levitation, two ranging from satellites to biomedical equipment [7]. Indeed, centering devices and a longitudinal AMB are necessary. magnetic bearings have many advantages over their conven- Obviously, the nature of the forces involved introduces tional counterparts: important nonlinearities in the physics of an axis. In addition, • Thanks to contactless, mostly frictionless operation, AMBs being very fast electromagnetic devices, major real- they can support loads with very high rotational speeds. time constraints have to be considered when designing an • Since they do not require lubrication, they are suitable appropriate control system. for environments where excluding contamination is key, Control of magnetic levitation systems, are the subject of such as clean rooms, or where efficient lubrication is a numerous publications owing to their industrial importance problem, such as deep vacuums. (see e.g [2]–[6], [10], [17], [19]–[21], [25], [28], [31]), which The extension of E ARNSHAW’s theorem to magnetic rely on a wide array of modern control techniques. What forces shows it is impossible to design stable positioning makes this control problem hard stems mainly from its systems by the mere use of permanent ferromagnetic mag- complex model. nets. While passive solutions based on diamanetic materials The purpose of this paper is thus to apply the “model- exist [23], they are uncommon in practice. This is why most free control” approach to that problem. Introduced by [11], applications implement active magnetic bearings (AMBs). [12], it has already been used successfully to solve numerous Active Magnetic Bearings are electromagnet-based and control problems spanning diverse application areas [14]. require an active control system to operate correctly [26]. Moreover, for each real studied cases, local approximation They operate as follows. Each control axis (see Figure 1) model could be first order. Specificities of magnetic bearings features two electromagnets and a position sensor measuring – most importantly negligible friction – introduce novelty rotor displacement. Each electromagnet generates a force from the point of view of possible local model choice and which is proportional to the square of its coil current, and allows us to contribute a new, simpler and more natural inversely proportional to the square of the air gap between formulation of that technique. its stator and the supported shaft. Through modulation of This paper is organized as follows. First, section II de- these forces, it is possible precisely position the shaft along scribes the new formulation of model-free control. Then, its the control axis. A centering device able to position a shaft application to magnetic bearings, including lab experiments and a performance comparison with two different control Jérôme De Miras (assistant professor) and Stéphane Bonnet (research techniques are discussed in section III. Finally, some insight engineer) are with Heudiasyc (CNRS, UMR 7253), Université de Technolo- into future developments is given in section IV. gie de Compiègne, BP 20529, 60205 Compiègne, France {demiras, sbonnet}@hds.utc.fr Samer Riachy (assistant professor) is with INRIA – Non- II. M ODEL - FREE CONTROL A & ECS-Lab (EA 3649), ENSEA, 95014 Cergy-Pontoise, samer.riachy@ensea.fr A. General principles Cédric Join (assistant professor) is with INRIA – Non-A & CRAN Model-free control is based on continually estimating a (CNRS, UMR 7039), Université de Lorraine, BP 239, 54506 Vandœuvre- lès-Nancy, France Cedric.Join@univ-lorraine.fr numerical difference between a expect local model (can Michel Fliess (CNRS research director) is with LIX (CNRS, be seen as a virtual linear tangent system) and the real UMR 7161), école polytechnique, 91128 Palaiseau, France behavior of a system from the sole knowledge of its input- Michel.Fliess@polytechnique.edu Cédric Join and Michel Fliess are with AL.I.E.N., 24-30 rue Lionnois, output signals. Let (1) be the unknown differential equation, BP 60120, 54003 Nancy, France www.alien-sas.com possibly nonlinear, describing the input-output behavior with Preprint submitted to 52nd IEEE Conference on Decision and Control . Received March 11, 2013. CONFIDENTIAL. Limited circulation. For review only. Remark 3: Setting KD = 0 in (2) yields an intelligent PI or iPI. This is by far the most common case in practice (see [14]). B. Estimation of F and effective construction of u 1) General remark: according to (2), estimating F is straightforward (see the following paragraph). Even if it may be necessary to denoise u and y, elementary filters have been sufficient until now. If KD 6= 0 in (2), estimating F requires an estimate of the derivative y,˙ i.e. the derivative of a noisy signal. Basic derivative filters are usually sufficient. Remark 4: The case of higher noises has been success- fully handled in practical settings [30] through algebraic techniques detailed in [13], [22]. Fig. 2. Control plane with two perpendicular axis 2) Used construction: according to equation (2), authors choose to construct Fˆ as follow Z ˆ F = −αu + KP e + KI e + KD e˙ ∗ (5) input u and output y with u∗ a denoised value of the control input. uk − (KP ek + KD D(z).ek ), (7) where α ⋆ • e = y − y is the difference between the output y and where z is the delay operator, N (z) a discrete filter applied to the desired output y ⋆ , the control input and D(z) discrete derivative operator. N (z) • KP , KI et KD are tuning parameters, function is twofold. It serves as a denoising filter for the • F , subsuming (1), must be numerically estimated in real control input and prevents the apparition of an algebraic loop time, in the control formulation. Ki is zero as explained below. As • α, non-physical parameter must be chosen by the prac- already pointed out by the authors [1], model-free control titioner so that F and αu have the same order of leads to very similar forms of classical PID controllers but magnitude. without having to use for setting a linear approximation of the system (at most the static gain). Let Fˆ be an estimate of F . The feedback control is chosen with the following form III. M AGNETIC BEARINGS A. A simplified model Fˆ u=− (3) The model used for simulations considers a single axis α and is thus monovariable. Focusing on axis y (Fig. 2), the Together, (2) and (3) are called intelligent PID controller radial acceleration of the rotor can be written as by the authors, or iPID because of the form of (2) and by analogy with [11], [12], [14]. It leads to m¨ y = Fyp + Fym + Fp , (8) where Fyp and Fym are the coil-generated magnetic forces Z KP e + KI e + KD e˙ = 0, (4) and Fp an additive, constant disturbance such as gravity. Neglecting the effects of magnetic saturation and hysteresis, which is stable for a suitable choice of gains. it follows Remark 1: This formulation does not require the discus- sion of the order of a local model, as in the previous λ1 ixp 2 λ2 ixm 2 Fyp = and F ym = − (9) propositions of the authors (see [11], [12], [14]). This is a 2(e0 − y)2 2(e0 + y)2 substantial advantage. where e0 is the nominal gap between the shaft and the coils Remark 2: If necessary (2) can be replaced by and parameters λ1 and λ2 depend on the electromagnet and F + αu − KP e − Φ(e) = 0, shaft geometries. Since each axis consists of two symmetrical actuators, the latter are both assumed equal to the single where Φ is a functional of e. Hence, model-free control parameter λy . Combining equations (8) and (9), yields a allows a wide array of possible control laws. model which is not linearizable at the origin – where the shaft Preprint submitted to 52nd IEEE Conference on Decision and Control . Received March 11, 2013. CONFIDENTIAL. Limited circulation. For review only. −4 x 10 1.5 3 Reference signal input Output 1 2 0.5 1 0 0 −0.5 −1 −1 −2 −1.5 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 −3 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 Fig. 3. Output (red) and reference (black) Fig. 4. Control signal u is centered and currents are zero (see [4]). However, a model suitable for linear analysis and control design can be obtained by applying a constant premagnetization bias current I0 to both coils. The constant magnetic flux present in the two actuators eliminates the flux creation time, which leads to an almost linear response of the shaft for small current variations around I0 . Using a bias current has one major flaw though. Since the two coils are always active, their energy consumption is much higher. Nonlinear operation of an AMB is thus more efficient, as only one of the coils is active at any time. In the latter operating mode, currents iyp and iym are mutually exclusive and can be expressed as a function of a single virtual current iy : −iy if iy < 0 iy if iy > 0 Fig. 5. The test-bench. iym = et iyp = 0 otherwise 0 otherwise. (10) Equation (10) implies that either Fyp , or Fym is zero at any given time. Equation (9) then yields λy sign(iy )iy 2 Fyp + Fym = , (11) 2(e0 − sign(iy )y)2 Filtering, sensors conditioning which gives the simulation model through substitution into Power (8). electronic B. Model parameters Values of the physical parameters m, λy and e0 are given D/A Conversion A/D Conversion in table I. The setpoint follows a low-pass-filtered, 10-Hertz Actuators Position square signal. As magnetic bearings are subject to minimal control measurements damping, the chosen control law is an iPD controller which, Real-time control software + xPC Target kernel from [1], matches a classical PID controller. Its parameters are KI = 0, α = .9, KP = 14692 and KD = 266. These PC Industriel values are the same as those of the nonlinear PID use Ethernet for comparison in real experiment. In the simulation, the constant disturbance Fp changes sign Software for programming and supervision at t = .25 s. A low amplitude noise (less than 2 × 10−6 ) is MATLAB, Simulink added to the output. The simulation results, obtain with the simplified model and the control law (7) given on Figures 3 Desktop PC and 4, show the efficacy of the control law without any tuning Fig. 6. Overview of the control system after the choose of the parameter values. Preprint submitted to 52nd IEEE Conference on Decision and Control . Received March 11, 2013. CONFIDENTIAL. Limited circulation. For review only. C. Experimentation 12 In contrast to the simulations presented above, the exper- Trajectory iment deals with a complete bearing where all degrees of 10 Response Reference signal freedoms are driven simultaneously by model-free control laws. 8 The test-bench used for these experiments is a laboratory Position (% of e0) AMB supplied by MECOS-TRAXLER AG, model miniVS 6 (Figure 5). It features a magnetic suspension unit comprising a rotor, two active centering devices and an active longitu- 4 dinal bearing, whose parameters are summarized in Table 2 I. 0 TABLE I 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Time (s) E XPERIMENTAL AMB PARAMETERS Fig. 7. y2 axis – Response to a square reference signal Parameter Variable Value Rotor mass m 3, 097 kg Maximum coil current Imax 6A −1.9 Nominal gap e0 0, 5 10−3 m Coil parameter λy,z 2, 51 10−6 mH.m −2 Acquisition period Tm 33 10−6 s except invMod control 66 10−6 s ADC resolution 12 − bit −2.1 ADC measurement range 0, 5 10−3 m Current (A) Input numeric filter to 1 10−3 s −2.2 time constant −6 Control laws Te 132 10 s period −2.3 −2.4 All five control axes are driven by a single PC running real-time Simulink code. Inputs are sampled at a frequency −2.5 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Time (s) higher than that of the control law to allow for efficient Fig. 8. y2 axis – Control signal filtering. Two series low-pass filters with a time constant to are used to this end. In order to assess the performance of the model-free ap- to a perturbation due to gravity, while the x axis is to be proach, a total of three control laws have been implemented: stabilized close to the nonlinearizable origin of the model1 . • The model-free control described above. All axes are As shown on Figures 13 and 14, keeping the x-axis at assumed independent. the origin is hard as it is the point where the coil currents • A global, Euler-Lagrange model-based nonlinear con- are zero. The time needed to establish a current in each coil trol law [9]. PID controllers are tuned to output desired induces a slight delay that prevents instantaneous reaction values for the second derivatives of the generalized co- from the controller. In contrast, this phenomenon does not ordinates of the model according to chosen closed-loop occur on the y and z axes since a nonzero current is always dynamics. The third order of this closed loop is form flowing through the coils to oppose gravity. as a product of a first order (time constant: 0.0045 s) Figure 7 shows the value of the y2 -axis position, featuring and a second order (angular frequency: 180 rd.s−1 and both the reference square signal and a desired output signal damping factor: 1.1). Full model equations are then used obtained through low-pass filtering of the former. The match- to compute the matching currents to apply to the plant. ing control signal is shown on Figure 8. A significant noise • A discrete nonlinear controller [3] where desired cur- level can be observed as the input filter does not completely rents are obtained through a table-based numerical cancel measurement noises and the derivative term D of the inversion of the behavior of an axis as a function of controller is a rough approximation. the desired shaft position at the next time step. The control signal itself is shaped by the combined influ- Let y1 , z1 and y2 , z2 be the positions of the shaft ends. ence of three elements: y1 , z1 and the x axis are kept at the nominal gap by an iPD • Since both y and z axes are directed towards the ground, controller. y2 is made to follow a square reference signal negatives currents are needed to compensate for gravity. varying from zero to e0 /8 at a frequency of 2 Hz. Likewise, • The value of the current necessary to compensate for z2 is made to follow a sinusoidal reference signal varying gravity is −2.07 A for the nominal gap. from −e0 /8 to e0 /8 at the same frequency. Authors chose 1 As the table on which the test-bench resides is not perfectly horizontal, this reference signal as good indicators to to interpret the this axis still experiences some gravity. This explains the non-zero average performance of control laws. Both y and z axes are subject of the control signal shown on Figure 14 Preprint submitted to 52nd IEEE Conference on Decision and Control . Received March 11, 2013. CONFIDENTIAL. Limited circulation. For review only. 01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 1.1 1 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 1.1 Time (s) Time (s) Fig. 9. y axis – Step response Fig. 10. Comparison between the three controllers on the y2 axis. 15 • The offset relative to the nominal gap. In the experi- 10 Trajectory Response Reference signal ment, the rotors moves away from the electromagnet which maintains its position. Hence, the current has 5 to increase with the gap to keep the magnitude of the force opposing gravity. The model error estimator Fˆ Position (% of e0) 0 perfectly fulfills its function and compensates for this nonlinearity. −5 Figure 9 details a single step response of the y2 -axis. The system response is slightly ahead of the reference signal y2⋆ −10 it should be following. Indeed, Fˆ compensates for errors between the real system and that on which the control law −15 1 1.2 1.4 1.6 1.8 2 2.2 2.4 is based with a slight delay. This phenomenon also depends Fig. 11. Time (s) z2 axis – Sinusoidal trajectory tracking on the value of the α parameter, here 2.02 for both the y and z axes. Figure 10 details a single step response of the y2 -axis of all three control laws. They all have been tuned to feature nonlinearity on currents is at its highest, it is hard to define an the same response time. optimal value for the α coefficient of the emphiPD controller Compared to the nonlinear global PID control, the model- in this case. Event if the model inversion-based controller free controller also eliminates the steady state error due better captures that nonlinearity, stabilization is still achieved to gravity but without any overshoot. Its behavior is also by the model-free controller, albeit a larger noise. almost indistinguishable from that of the model inversion- based controller. Moreover, it achieves this result with a IV. C ONCLUSION AND FURTHER WORK much lighter computing cost, keeping in mind the model The model-free controller yields as good results than the inversion-based controller had been giving the best results nonlinear PID (The difference is that the PID has not filtered on this test-bench until the present experimentation. reference of the step signal of the input) and matches the Figures 11 and 12 show the z2 axis response to its laboratory benchmark model inversion-based controller with sinusoidal reference and the matching control signal. The a much lower computing cost than the two other control laws. results are again correct as expected. Still, the level of noise obtained when trying to stabilize the Finally, Figures 13 and 14 show the stabilization perfor- shaft a a point where the current is zero and the model is mance of the controller for the x axis and the associated nonlinearizable shows a tight coupling of the quality of the control signal. The high frequency content of the control results and the α tuning parameter. This shows a possible signal is the result of the noise generated by the derivative path towards improvements of this control approach. As the approximation use in the PI function with a little sampling proposed control scheme is not directly connected to the period and minimum interval measurement of positions (see AMB model, it remains valid as well for this type of devices Table I). at different scales than other systems. Figure 13 also shows the response of the model inversion- based controller, which yields a tighter stabilization of the R EFERENCES shaft at the origin, which is nonlinearizable in the model [1] B. d’Andréa-Novel, M. Fliess, C. Join, H. Mounier, B. Steux (2010). as explained before. Since at this point the impact of the A mathematical explanation via “intelligent” pid controllers of the strange ubiquity of pids. 18th Medit. Conf. Control Automat., 2 The difference with the value used in simulations is due to a heavier Marrakech. Accessible at real system http://hal.archives-ouvertes.fr/inria-00480293/fr/. Preprint submitted to 52nd IEEE Conference on Decision and Control . Received March 11, 2013. CONFIDENTIAL. Limited circulation. For review only. −1.6 0.1 −1.7 0 −1.8 −0.1 −1.9 Current (A) Current (A) −0.2 −2 −0.3 −2.1 −0.4 −2.2 −0.5 1 1.2 1.4 1.6 1.8 2 2.2 2.4 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Time (s) Time (s) Fig. 12. z2 axis – Control signal Fig. 14. x axis – Control signal 0.5 Trajectory Response 0.4 Reference signal easy. Int. J. Model. Identif. Control, vol. 4, p. 12–27. Accessible at InvMod response 0.3 http://hal.archives-ouvertes.fr/inria-00158855/fr/. [14] M. Fliess, C. Join, S. Riachy (2011). Rien de plus utile qu’une 0.2 bonne théorie: la commande sans modèle. JD-JN MACS, Marseille. Accessible at 0.1 http://hal.archives-ouvertes.fr/hal-00581109/fr/. Position (% of e0) 0 [15] M. Fujita, F. Matsumara, M. Shimizu (1990). H ∞ robust control design for a magnetic suspension system. 2nd Int. Symp. Magnetic −0.1 Bearings, Tokyo. [16] W. Grega, A. Pilat (2005). Comparison of linear control methods for −0.2 an amb system. Int. J. Appl. Math. Comput. Sci., vol. 15, p. 245–255. −0.3 [17] T.R. Grochmal, A.F. Lynch (2007). Control precision tracking of a ro- tating shaft with magnetic bearings by nonlinear decoupled disturbance −0.4 observers. IEEE Trans. Control Syst. Techno., vol. 15, p. 1112–1121. −0.5 [18] J. Huang, L. Wang, Y. Huang (2007). Continuous time model 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Time (s) predictive control for a magnetic bearing system. PIERS Online, vol. Fig. 13. x axis – stabilization at zero 3, p. 202–208. [19] M.S. Kang, W.H. Yoon (2006). Acceleration feedforward control in active magnetic bearing system subject to base motion by filtered-x lms algorithm. IEEE Trans. Control Syst. Techno., vol. 14, p. 134–140. [2] H.M.N.K. Balini, C.W. Scherer, J. Witte (2010). Performance en- [20] J. Lévine, J. Lottin, J.-C. Ponsart (1996). A nonlinear approach to the hancement for amb systems using unstable H∞ controllers. IEEE control of magnetic bearings. IEEE Trans. Control Syst. Techno., vol. Trans. Control Syst. Techno., vol. 19, p. 1479–1492. 4, p. 524–544. [3] S. Bonnet, J. De Miras, B. Vidolov (2008). Nonlinear one-step [21] F. Mazenc, M. S. de Queiroz, M. Malisoff, F. Gao (2006). Further predictive control of an active magnetic bearing. 17th IFAC World results on active magnetic bearing control with input saturation. IEEE Congress, Seoul. Trans. Control Syst. Techno., vol. 14, p. 914–919. [4] A. Charara, J. De Miras, B. Caron (1996). Non-linear control of a [22] M. Mboup, C. Join, M. Fliess (2009). Numerical differentiation with magnetic levitation system without premagnetization. IEEE Trans. annihilators in noisy environment. Numer. Algor., vol. 50, p. 439–467. Control Syst. Techno., vol. 4, p. 513–523. [23] R. Moser, J. Sandtner, H. Bleuler (2001). Diamagnetic suspension system for small rotors. J. Micromechatronics, vol. 1, p. 131–137. [5] S. Chen (2011). Robust nonlinear smooth feedback control of a [24] K. Nonami, H. Yamaguchi (1994). µ-synthesis of a flexible rotor three-pole active magnetic bearing system. IEEE Trans. Control Syst. magnetic bearing system. 4th Int. Symp. Magnetic Bearings, Zurich. Techno., vol. 19, p. 615–621. [25] M.S. de Queiroz, S. Pradhananga (2007). Control of magnetic [6] S.-Y. Chen, F.-J. Lin (2011). Robust nonsingular terminal sliding- levitation systems with reduced steady-state power losses. IEEE Trans. mode control for nonlinear magnetic bearing system. IEEE Trans. Control Syst. Techno., vol. 15, p. 1096–1102. Control Syst. Techno., vol. 19, p. 636–643. [26] G. Schweitzer, H. Bleuler, A. Traxler (1994). Active Magnetic [7] A. Chiba, D.G. Dorrell, T. Fukao, O. Ichikawa, M. Oshima, M. Take- Bearings. VDF Hochschulverlag. moto (2005). Magnetic Bearings and Bearingless Drives. Elsevier. [27] H.F. Steffani, W. Hofmann, B. Cebulski (1998). A controller for a [8] D. Cho, Y. Kato, D. Spilman (1993). Sliding mode and classical magnetic bearing using the dynamic programming of Bellman. 6th controllers in magnetic levitation systems. IEEE Control Systems International Symposium on Magnetic Bearings, MIT, Cambridge, Magaz., vol. 13, p. 42–48. MA. [9] J. De Miras, A. Charara (1998). A vector oriented control for a [28] N. Van Dijk, E. Van De Wouw, N. Doppenberg, H. Oosterling, H. magnetically levitated shaft. IEEE Trans. Magnetics, vol. 34, p. 2039– Nijmeijer (2010). Robust active chatter control in the high-speed 2041. milling process. Amer. Control Conf., Baltimore. [10] H. Du, N. Zhang, J.C. Ji, W. Gao (2010). Robust fuzzy control of [29] B. Vidolov, C. Melin, J. De Miras, A. Charara (1996). Two-rules-based an active magnetic bearing subject to voltage saturation. IEEE Trans. fuzzy logic control and sliding mode control of an active magnetic Control Syst. Techno., vol. 18, p. 164–169. bearing. 5th Int. Conf. Fuzzy Systems, New Orleans. [11] M. Fliess, C. Join (2008). Commande sans modèle et commande à [30] J. Villagra, B. d’Andréa-Novel, M. Fliess, H. Mounier (2011). A modèle restreint. e-STA, vol. 5 (n◦ 4), p. 1–23. Accessible at diagnosis-based approach for tire-road forces and maximum friction http://hal.archives-ouvertes.fr/inria-00288107/fr/. estimation. Control Eng. Practice, vol. 19, p. 177–184. [12] M. Fliess, C. Join (2009). Model free control and intelligent pid [31] K. Youcef-Toumi, S. Reddy (1992). Dynamic analysis and control of controllers: towards a possible trivialization of nonlinear control? 15th high speed and high precision active magnetic bearing. J. Dyn. Sys. IFAC Symp. System Identif., Saint-Malo. Accessible at Meas. Control., vol. 114, p. 623–633. http://hal.archives-ouvertes.fr/inria-00372325/fr/. [13] M. Fliess, C. Join, H. Sira-Ramírez (2008). Non-linear estimation is Preprint submitted to 52nd IEEE Conference on Decision and Control . Received March 11, 2013.

US$309.95 Free Express Shipping Download and try the fully functional Multi-Instrument software using your sound card as the ADC and DAC device! 1. Instroduction This is one of the second-generation USB DSO (USB Oscilloscopes) designed and developed by Virtins Technology. This generation of USB DSOs features Virtins Technology’s unique hardware-based DSP algorithm which enhances the performance and functionality dramatically without adding extra hardware cost. When used in conjunction with Multi-Instrument® software, the USB DSO converts any desktop, laptop, or tablet PC into a powerful oscilloscope, spectrum analyzer, multimeter, data logger, signal generator and so forth, all of which work simultaneously. 2. Package Contents 1) VT DSO-2820 unit with a hardware bundled Multi-Instrument Standard software license 2) 2 × 100MHz Oscilloscope Probe P6100 with two switchable positions: × 1, × 10 3) Signal Generator Test Lead (1 m) 4) USB cable (1.5 m) 5) CD (contains the copy-protected Multi-Instrument Software and VT DSO-2820 driver) 6) Individual voltage calibration data |
| Oscilloscope Horizontal (Time) Axis | |||
|---|---|---|---|
| Real Time Sampling Frequency (fs) | Single Analog Channel 200 MHz Single or Dual Analog Channels with or without 1-bit Digital Channel 100 MHz, 50 MHz, 25 MHz, 20 MHz, 10 MHz, 5 MHz, 4 MHz, 2 MHz, 1 MHz, 500 kHz, 400 kHz, 200 kHz, 100 kHz, 50 kHz, 40 kHz, 20 kHz, 10 kHz, 5 kHz, 4 kHz, 2 kHz, 1 kHz, 500 Hz, 400 Hz, 200 Hz, 100 Hz, 50 Hz, 40 Hz, 20 Hz, 10 Hz, 5 Hz, 4 Hz, 2 Hz, 1 Hz | ||
| Equivalent Sampling Frequency | = [Real Time Sampling Frequency] × [Number of Frames Persisted]. Maximum 40 GHz, for repetitive signals whose maximum frequency is less than ¼ of the real time sampling frequency. Not valid for the case of post-trigger. | ||
| Buffer Size | Normal Frame Mode | Analog | Single Channel 40000 samples (8 bits) 20000 samples (16 bits) Dual Channels 20000 samples / channel (8 bits, Non-ALT mode) 10000 samples / channel (16 bits, Non-ALT mode) 40000 samples / channel (8 bits, ALT mode) 20000 samples / channel (16 bits, ALT mode) |
| Digital | 20000 samples (1 bit) | ||
| Record Mode (Streaming Mode) | Limited only by hard disk space available and maximum file size allowed by the operating system. Maximum sampling frequency for continuous streaming is computer speed and software setting dependent and typically 10 MHz (single channel, 8 bit) in Multi-Instrument. | ||
| Roll Mode (Streaming Mode for Low Frequency Signals) | Limited only by the computer memory available. Roll Mode is allowed when fs ≤ 1MHz and [Record Length] ≥ 4 × [Roll Width]. Maximum sampling frequency for continuous streaming is computer speed and software setting dependent. | ||
| Sweep Time | 5 ns ~ 500 s (Non-Streaming Mode) | ||
| Sampling Frequency Accuracy | ± 50 ppm | ||
| Oscilloscope Vertical (Analog) Axes | |||||||
|---|---|---|---|---|---|---|---|
| Number of Channels | 2 (i.e. Ch. A and Ch. B) | ||||||
| ADC Bit Resolution | 8 Bits | ||||||
| Enhanced ADC Bit Resolution (available only when Sampling Frequency is less than 100 MHz) | 16 Bits If this option is selected, the effective bit resolution increases from 8 bits to up to 16 bits as the sampling frequency goes down. | ||||||
| Sampling Frequency | Effective Bit Resolution | Sampling Frequency | Effective Bit Resolution | ||||
| ≥ 100 MHz | 8 Bits | ≤ 25 MHz | 9 Bits | ||||
| ≤ 6.25 MHz | 10 Bits | ≤ 1.563 MHz | 11 Bits | ||||
| ≤ 391 kHz | 12 Bits | ≤ 97.7 kHz | 13 Bits | ||||
| ≤ 24.4 kHz | 14 Bits | ≤ 6.10 kHz | 15 Bits | ||||
| ≤ 1.526 kHz | 16 Bits | ||||||
| Bandwidth | fs > 50 MHz | 80 MHz | |||||
| fs ≤ 50 MHz | 8 Bits | No Effective Bit Resolution Enhancement | 80 MHz | ||||
| Effective Bit Resolution Enhancement | 15 ~ 25 MHz | ||||||
| 16 Bits | Effective Bit Resolution Enhancement | about 0.443 fs | |||||
| Voltage Measuring Range (Full Scale) | ± 10 mV, ± 20 mV, ± 50 mV, ± 100 mV, ± 200 mV, ± 500 mV, ± 1 V, ± 2 V, ± 5 V, ± 10 V, ± 20 V, ± 50 V | ||||||
| Max. Allowed Voltage | ± 100 V (DC + AC peak), derated above 100kHz | ||||||
| DC Accuracy | ± 1% | ||||||
| Coupling Type | AC / DC | ||||||
| Input Isolation | No (Isolation can be achieved through a USB isolator) | ||||||
| Terminal Type | Referenced Single-Ended, BNC | ||||||
| Input Impedance | 1 MΩ , 15 pF | ||||||
| Zero Calibration | Through hardware. Individually done at factory, user adjustable | ||||||
| Gain Calibration | Through hardware. Individually done at factory, user adjustable | ||||||
| Number of Channels | 1 (i.e. External Trigger Channel, 1-bit ADC) |
| Bandwidth | 140 MHz |
| Threshold Resolution | 45 mV |
| Threshold Hysteresis | 225 mV |
| Threshold Range | ± 20 V |
| Max. Allowed Voltage | ± 100V (DC + AC peak), derated above 100kHz |
| Threshold DC Accuracy | ± 1% |
| Coupling Type | DC |
| Input Isolation | No (Isolation can be achieved through a USB isolator) |
| Terminal Type | Referenced Single-Ended, BNC |
| Input Impedance | 1 MΩ , 15 pF |
| Zero Calibration | Through software. Individually done at factory. |
| Gain Calibration | Through software. Individually done at factory. |
| Trigger Detection Method | Digital |
| Trigger Source | Ch. A, Ch. B, EXT, ALT |
| Trigger Mode | Auto, Normal, Single, Slow |
| Trigger Edge | Rising, Falling |
| Trigger Level | Adjustable within full scale |
| Pre-Trigger | -100% ~ 0% of Record Length |
| Post-Trigger | 0 ~ 100% of Record Length |
| Trigger Frequency Rejection | Nil: No Rejection HFR: High Frequency Rejection, cut off at 0.11fs NR0: Noise Rejection, hysteresis = 1% of half of full scale NR1: Noise Rejection, hysteresis = 2% of half of full scale NR2: Noise Rejection, hysteresis = 4% of half of full scale NR3: Noise Rejection, hysteresis = 8% of half of full scale NR4: Noise Rejection, hysteresis = 16% of half of full scale HN0: HFR + NR0 HN1: HFR + NR1 HN2: HFR + NR2 HN3: HFR + NR3 HN4: HFR + NR4 HNX: selectable HFR + adjustable hysteresis = 0% ~ 25% of half of full scale Note: The specified hysteresis may be modified internally to ensure [Trigger Level (%)] – [Hysteresis (%)] ≥ -100% at rising edge, or [Trigger Level (%)] + [Hysteresis (%)] ≤ 100% at falling edge. |
| THD | fs=100 kHz, f=1 kHz, from 2nd to 20th order, full-scale input: 8 bits (without bit resolution enhancement): ≤ -55 dB 8 bits (with bit resolution enhancement): ≤ -56 dB 16 bits: ≤ -59 dB |
| IMD (250 Hz + 8 kHz, 4:1) | fs=100 kHz, from 2nd to 3rd order, full-scale input: 8 bits (without bit resolution enhancement): ≤ -51 dB 8 bits (with bit resolution enhancement): ≤ -52 dB 16 bits: ≤ -57 dB |
| IMD (19 kHz +20 kHz, 1:1) | fs=100 kHz, 2nd order only, full-scale input: 8 bits (without bit resolution enhancement): ≤ -65 dB 8 bits (with bit resolution enhancement): ≤ -65 dB 16 bits: ≤ -78 dB |
| SFDR | fs=100 kHz, f=1 kHz, full-scale input: 8 bits (without bit resolution enhancement): ≥ 61 dB 8 bits (with bit resolution enhancement): ≥ 61 dB 16 bits: ≥ 62 dB |
| Crosstalk | ≤ -45 dB (at the same voltage measuring range for full bandwidth) |
| Noise | For voltage measuring ranges ± 50 mV and above: 8 bits (without bit resolution enhancement): ≤ ± 3 counts (± 1%) 8 bits (with bit resolution enhancement, fs =100 kHz): ≤ ± 2 counts (± 0.8%) 16 bits (fs =100 kHz): ≤ ± 0.3 8-bit count (± 0.1%) For voltage measuring ranges ± 10 mV and ± 20 mV: 8 bits (without bit resolution enhancement): ≤ ± 15 counts (± 6%) 8 bits (with bit resolution enhancement, fs =100 kHz): ≤ ± 9 counts (± 4%) 16 bits (fs =100 kHz): ≤ ± 1 8-bit count (± 0.4%) |
| Number of Channels | 1 |
| Coupling Type | DC |
| Output Isolation | No |
| Terminal Type | Referenced Single-Ended, BNC |
| Output Impedance | About 600 Ω |
| Overvoltage Protection | ± 35 V |
| Output Voltage Range | ± 2 V, adjustable | |
| DAC Bit Resolution | 10 Bits | |
| Output Sampling Frequency (fs) | DDS Mode or Streaming Mode 6.25 MHz, 3.125 MHz, 1.5625 MHz, 625 kHz, 312.5 kHz, 125 kHz, 62.5 kHz, 31.25 kHz, 25 kHz, 12.5 kHz, 6.25 kHz, 3.125 kHz | |
| Output Sampling Frequency Accuracy | ± 50 ppm | |
| Bandwidth | DC ~ 150 kHz | |
| Output Signal Frequency | 0 ~ ½ of output sampling frequency | |
| Rise Time (10% ~ 90%) | < 2.3 µs | |
| DC Offset Range | Full output voltage range | |
| DC Accuracy | ± 0.5% of full scale | |
| Waveform | DDS Mode* | Sine, Rectangle (duty cycle adjustable), Triangle, Saw Tooth, White Noise, MLS (length = 263-1), User Configurable Waveform Library (Arbitrary), Musical Scale |
| Streaming Mode* | Sine, Rectangle (duty cycle adjustable), Triangle, Saw Tooth, White Noise, Pink Noise, MultiTones, MLS (127~16777215), DTMF, User Configurable Waveform Library (Arbitrary), Musical Scale | |
| Signal Frequency Resolution | DDS Mode | <0.0007 Hz (fs = 3.125 MHz) |
| Streaming Mode | Virtually infinitesimal | |
| Buffer Size | DDS Mode | Without interpolation: 1024 samples With interpolation: 1024 × 65536 = 67108864 samples (Note: interpolation is only available when fs <= 3.125 MHz) |
| Streaming Mode | Virtually unlimited | |
| Frequency Sweep | DDS Mode | Supports linear sweep of all types of repetitive waveforms. Sweep speed range: 1/32768 × fs2 / 232 ~ 65535 × fs2 / 232 (e.g. 0.017 Hz/s ~ 37.3 MHz/s when fs = 1.5625 MHz) |
| Streaming Mode | Supports linear and logarithmic sweep of all types of repetitive waveforms. Sweep speed range: unlimited | |
| Amplitude Sweep | DDS Mode | Supports linear sweep of all types of waveforms. Sweep speed range: 1/32768 × fs / (232 –1) ~ 65535 × fs / (232 –1) (e.g. 0.00000111 %/s ~ 2384 %/s when fs = 1.5625 MHz) |
| Streaming Mode | Supports linear and logarithmic sweep of all types of waveforms. Sweep speed range: unlimited | |
| Duration (Signal Length) Resolution | DDS Mode | 1/fs or 1µs, whichever is greater |
| Streaming Mode | 1/fs | |
| THD | ≤ -60 dB (fs = 1.5625 MHz, f = 1 kHz, from 2nd to 20th order, full-scale output) | |
| SFDR | ≥ 65 dB (fs = 1.5625 MHz, f = 1 kHz, full-scale output) | |
| Zero Calibration | Through software. Individually done at factory. | |
| Gain Calibration | Through software. Individually done at factory. | |
| Voltage Range | 0 ~ 3.3 V, not adjustable | ||
| Output Signal Frequency Accuracy | ± 50 ppm | ||
| Bandwidth | 13 MHz | ||
| Rise Time (10% ~ 90%) | < 50 ns | ||
| Waveform | Square | Signal Frequency | 25 MHz / N, (N=1, 2, 3, ….25000) |
| MLS | Sampling Frequency | 25 MHz / N, (N=1, 2, 3, ….25000) | |
| Interface | USB 2.0 High Speed / USB 1.1 Full Speed / USB Isolator | |
| Device Category in Multi-Instrument | ADC Device | VT DAQ 1 |
| DAC Device | VT DAO 1 | |
| Firmware Upgradable | Yes | |
| Power | Bus powered by USB port, no external power source required. | |
| Power Consumption | Max. 2W | |
| Dimensions | 145 mm (L) × 108 mm (W) × 26 mm (H), anodized aluminum case | |
| System Requirement | Windows XP, Vista, 7, 8 or above, 32 bit or 64 bit | |
| Operating Temperature | 0° C ~50° C | |
| Attenuation Ratio | × 1, × 10 |
| Bandwidth | DC ~ 100 MHz (× 10), DC ~ 6 MHz (× 1) |
| Input Impedance | 1 MΩ (× 1, with VT DSO connected) 10 MΩ (× 10, with VT DSO connected) |
| Input Capacitance | 18.5 pF ~ 22.5 pF (× 10), 85 pF ~ 120 pF (× 1) |
| Input Capacitance Compensation Range | 10 ~ 30 pF |
| Length | 1.2 m |



G2CCD cameras are designed to satisfy both amateur and professional astronomers. They are compact to be attached even to small telescopes. USB computer interface ensures plug-and-play operation. Camera construction is robust to sustain every-day usage. Usage of scientific-grade NABG CCD chips with high QE and linear response enables reliable scientific data reduction. Employing of 16-bit A/D converter with correlated double-sampling and low-noise electronics ensures high dynamic range. CCD chip can be cooled up to 35 °C below ambient temperature to enable long exposures. Internal shutter and filter wheel allow integration into robotic setups performing unattended observations...
G2CCD-0400 on small 80mm short-focus retractor
Please note the G2 CCD cameras are designed to work in cooperation with a host Personal Computer (PC). As opposite to digital still cameras, which are operated independently on the computer, the scientific slow-scan, cooled cameras usually require computer for operation control, image download, processing and storage etc. To operate G2 CCD camera, you need a computer which:
Is compatible with a PC standard.
Runs a modern 32-bit Windows operating system.
Provides at last one free USB port.
G2 cameras use high sensitivity and low noise Kodak KAF Full Frame CCD detectors. Advanced techniques like transparent electrodes and microlensing ensure up to 85 % peak quantum efficiency.
| CCD Chip | Kodak KAF-0402ME Class 1 or 2 |
| Resolution | 768 (H) × 512 (V) pixels |
| Pixel size | 9 um (H) × 9 um (V) |
| Imaging area | 6.9 mm (H) × 4.6 mm (V) |
| Full well capacity | ~100,000 e- |
| Output node capacity | ~220,000 e- |
| Read noise | 15 e- RMS |
| Thermal noise | 1 e-/s/pixel at 0 °C |
| Dark signal doubling | 6.3 °C |
CCD chip specifications
KAF-0402ME CCD chip and its Quantum Efficiency
16-bit A/D converter with correlated double sampling ensures high dynamic range and CCD chip-limited readout noise. Fast USB interface ensures image download time within seconds. Maximum length of single USB cable is 5 m. This length can be extended for instance to 10 m by using single USB hub or up to 100 m by third-party extender.
| ADC resolution | 16 bits, no missing codes |
| Sampling method | Correlated double sampling |
| Gain | 1.5 e-/ADU (1 × 1 binning) |
| 2.5 e-/ADU (other binnings) | |
| Horizontal binning | 1 to 4 pixels |
| Vertical binning | 1 to 4 pixels |
| Sub-frame readout | Arbitrary sub-frame |
| TDI readout | Yes, 2 ms resolution |
| System read noise | 15 e- RMS (low-noise read) |
| 16.5 e- RMS (standard read) | |
| Full frame download | 1.9 s (low-noise read) |
| 1.7 s (standard read) | |
| Computer interface | USB 1.1 Full Speed |
Camera electronics specifications
Notes:
Binning can be combined independently on both axes
Download times are valid for USB 1.1 electronics and may vary depending on host PC. USB 2.0 electronics is under development and download times will be different. USB 2.0 digital module will be available in Q1 2006.
USB 1.1 digital electronics module can be easily replaced by USB 2.0 version in existing cameras due to modular electronics design. Refer to our web site for upgrade cost.
Power and USB connectors on the bottom of the camera head
Regulated two-stage thermo-electric chip cooling up to °C35 °C below ambient temperature with forced air cooling and °C 0.1 °C temperature precision ensure very low dark current for long exposures.
| CCD chip cooling | Thermoelectric (Peltier modules) |
| TEC modules | Two stages |
| Max. delta T | 35 °C below ambient |
| Regulation precision | 0.1 °C |
| Hot side cooling | Air cooling (two 40 mm fans) |
Chip cooling specifications
Notes:
Water-assisted cooling will be available in 2006. Contact us for details.
The cooling performance depends on the environmental conditions and also on the power supply. If the power supply voltage drops below 12 V, the maximum temperature drop can be around 30 °C.
Two fans ensure effective cooling of the Peltier hot side
The 12 V DC power supply adapter enables camera operation from arbitrary power source including batteries, wall adapters etc. Universal 100–240 V AC/50–60 Hz, 60 W brick adapter is supplied with the camera. Although the camera power consumption does not exceed 30 W, the 60 W power supply ensures noise-free operation.
| Camera head supply | 12 V DC |
| Camera head power consumption | 30 W |
| Adapter input voltage | 100-240 V AC/50-60 Hz |
| Adapter output voltage | 12 V DC/5 A |
| Adapter paximum power | 60 W |
Power supply specifications
Notes:
Maximum power consumption includes 100% cooling.
The camera contains its own power supplies inside, so it can be powered by unregulated 12 V DC power source—the input voltage can be anywhere between 10 and 14 V. However, some parameters (like cooling efficiency) can degrade if the supply drops below 11 V.
G2 CCD camera measures its input voltage and provides it to the control software. Input voltage is displayed in the Cooling tab of the CCD Camera control tool in the SIMS. This feature is important especially if you power the camera from batteries.
12 V DC/5 A power supply adapter for G2CCD Camera
Warning:
The power connector on the camera head uses center-plus pin. Although all modern power supplies use this configuration, always make sure the polarity is correct if you use own power source.
Compact and robust camera head measures only 114 × 114 × 74 mm (approx. 4.5 × 4.5 × 3 inches). The head is CNC-machined from high-quality aluminum and black anodized. The head itself contains USB-B (device) connector and 12 V DC power plug, no other parts (CPU box, USB interface, etc.), except a brick power supply, are necessary. Integrated mechanical shutter allows streak-free image readout, as well as automatic dark frame exposures, which are necessary for unattended, robotic setups. Integrated filter wheel contains 5 positions for standard 1.25" threaded filter cells. A variant of filter wheel with 6 positions for 1" glass-only filters is also available.
| Internal mechanical shutter | Yes, blade shutter |
| Shortest exposure time | 0.1 s |
| Longest exposure time | Limited by chip saturation only |
| Internal filter wheel | 5-positions for 1.25" threaded filter cells |
| 6-positions for 1" bolt-secured filters | |
| Head dimensions | 114 mm × 114 mm × 74 mm (7 mm filter space) |
| 114 mm × 114 mm × 77 mm (10 mm filter space) | |
| Back focal distance | 25 mm (7 mm filter space) |
| 28 mm (10 mm filter space) | |
| Camera head weight | 1.1 kg |
Mechanical specifications
Notes:
Camera head weight with filter wheel and 1.25" nosepiece, but without filters.
Various telescope/lens adapters are available, including standard 1.25" and 2" barrels, T-thread adapter, Zeiss M44 × 1 thread adapter etc.
Integrated blade shutter and filter wheel (left) Custom CNC-machined heat sink with two high-efficient fans (right)
Integrated 5-positions filter wheel accommodates standard 1.25" filters (left) Scientific-grade BVRI filter set (right)
Powerful SIMS (Simple Image Manipulation System) software supplied with the camera allows complete camera control (exposures, cooling, filters) with automatic sequences and complete image calibration. SIMS also supports advanced tools like Image Add tool with automatic sub-pixel image alignment, (L)RGB Add tool, Image Blink tool, image filters and many more functions.
Simple Image Manipulation System
G2 CCD cameras are supplied in the foam-filled, hard carrying case containing:
Camera body with a user-chosen telescope adapter. The standard 1.25" barrel adapter is included by default. If ordered, the filter wheel is already mounted inside the camera head and filters are threaded into place (if ordered).
A 100-240 V AC input, 12 V DC output brick adapter with 1.8 m long power cable.
5 m long USB A-B cable for connecting camera to host PC.
A CD-ROM with camera drivers, SIMS software package with electronic documentation and PDF version of this manual.
A printed copy of this manual.
G2CCD camera in the carrying case
![]() Produkt kan variere for aktuelt produkt | Increase distance and solve your compatibility issues.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Share ![]()
3. VT DSO-2A20E Hardware Specifications
4. P6100 Oscilloscope Probe Hardware Specifications
5. Examples 1) Measurements of PAL Composite Video Signals ![]() 2) Measurements of a 500Hz sine wave and a 1kHz square wave under ALT trigger mode ![]() 3) Mixed signal display. Channels A & B are analog input channels while Channel EXT is used as a digital input channel ![]() Dual-Link DVI Repeater
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
0 Comments