I'm a postgraduate student at the Indian Institute of Astrophysics, Bangalore, India. I'm working on a 6- chaannel photodiode amplifier, which will be used for site surveys for a new telescope coming up in India.
A dual JFET opamp (LT1464) powered by ±15V supply is used to amplify the signals from a very sensitive photodiode (Thorlabs FDS1010) In the first stage, the current in the range of nanoamperes from the photodiode is converted to desirable volt range by a resistance of 9.1MΩ.
Thereafter, DC and AC components of the signal are separated and given different amplification coefficients with the second stage, along with another low pass filter that eliminates all frequencies above 400 Hz. AC component of the signal is amplified by 110 times, while the DC is amplified by 2 times so as to bring out the scintillations from what would have been a DC dominated signal. A high pass filter is also integrated into the circuit to only admit all frequencies over 0.1 Hz. Necessity of clean power requires appropriate smoothening capacitors, to be connected across the power supply of the amplifier
With the current circuit configuration, the circuit tends to oscillate giving an output sinusoidal signal of less than 15V amplitude, without the application of any signal. According to Fig, it’s giving a sinusoidal signal of about 10-12V with NO DC offset, and a frequency of 564Hz in the absence of light from the LED (or for that matter, any light at all). Even when the photodiode was disconnected, the oscillations were still present, as shown in the figure
Following is the PCB layout
After reading more about PCB design, I found that the PCB was full of parasitic capacitances and there was plenty of crosstalk between the output of the six channels. So I decided to separate all the channels and adopt a better PCB design
I'm making another board after reading through some PCB guidelines. I'm populating it only with SMD Resistors and caps (hopefully reducing the noise) and giving a proper ground and power plane. So, can you tell me whether there's any problem with this design before I give it for fabrication?
Following is the PCB design without the ground plane (for better visibility of the layout)
I have a startup and we have NO CLUE about any of this and we need something that is small, can play video or a photo slideshow without an OS and can attach a screen and need it as thin/compact as possible... How do we go about doing this and cheap or is their something that already exists?
I think you should better displace components, reducing trace lengths and loop areas. For example move 22pF and 15pF more closely to the opamp and change the routing of the connection between 10uF and 200K, avoiding the big loop around J3. Try also with traces mainly in one direction on one pcb side and perpendicular on the other one. In order to reduce parasitic capacitance, you should design pcb avoiding traces running too close on the same side or too close between the two pcb sides. In such critical circuit, pcb must be designed with care in order to avoid big issues.
Hope this help.
My project involves displaying time coming in computer clock on lcd by using AT 89c51 microcontroller.
Im not able to program 8951 properly .I need assembly lang program to display clock on LCD .A signal will come from computer DB -9 port to input port of 8051 via RS -232 cable ..
This project is way beyond my competency .So please help me completing this project
PLEASE HELP ASAP
please let me know on my email id : email@example.com
How thick was the pcb meant to be ?
Did the supplier agree the thickness tolerance when you ordered ? (in which case you could get the supplier to start again)
(Generally speaking I don't think that suppliers of prototype boards keep ultra careful control of board thickness - it's not a key issue to most users.)
If the board is too thin can't you pad it yourself ?
I acquired some copper clad boards at a good price (free). I then realized that these boards need to be prepped before I go through the UV process. I am currently using a positive process with a kit from MG Chemicals, so I assume I would need to use a coating for the positive process.could you advise me as to what chemical to use and where to get it? Also when we reach the point where we need commercial boards, we will be bugging you people... We are about a 40 min. drive from you in Vancouver WA.