Dear Dr. Stefan Ritt,
Thank you for your reply. My previous description was not accurate enough. We plan to modify the design referring to the attached file, and the details are as follows:
We will remove capacitor C2 to make the input DC-coupled. The signal ranging from -1 V to +1 V comes in via J1, then passes through a π-type attenuator to halve the signal amplitude before entering the amplifier. The amplifier gain is set to 2.
We are considering two dual supply options for the THS4508: +3.5 V / -1.5 V or +4 V / -1 V. The output common-mode voltage is set to 1.8 V through the CMOFS pin. After that, the signal goes through a voltage divider with 28 Ω and 22 Ω resistors.
The final signal to DRS4 has a common-mode voltage of about 0.8 V and a differential swing of -440 mV ~ +440 mV, which meets the DRS4 input requirements.
Could you please check if this THS4508 design is reasonable? We would also be grateful for any better suggestions or alternative solutions.
Thanks a lot for your support!
Best regards,
Mingxin Liu
| Stefan Ritt wrote: |
|
If you DC-couple the THS4508, you risk that it's output goes to the negative rail (in your case -1.5), which would destroy the DRS4 since it's out of its specs.
To move negative signals into the posive range of 0V to 2.5V (allowed range for DRS4), the only thing I see is AC coupling with properly biasing. I don't understand why you try to do DC coupling, that will not work. Alternatively, you coudl invert your negative signal with a small inverter (usually two passive coils) which would bring a negative signal into a positive range.
Best regards,
Stefan
| Mingxin Liu wrote: |
|
Dear Dr. Stefan Ritt,
Thank you so much for your detailed and helpful reply. It’s very clear to me now why the standard DRS4 design uses AC coupling, and I appreciate your practical suggestions on biasing and protection.
I have a follow-up question about the THS4508 amplifier I plan to use at the input stage, if you don’t mind. If I want to make the signal path DC-coupled (both input and output), would it be possible to power the THS4508 with dual supplies, e.g., +3.5 V and -1.5 V, so that the input can accept signals in the range of -1 V to +1 V? In this configuration, could the amplifier still achieve a bandwidth of 700–800 MHz, or is that high bandwidth only guaranteed when using a single +5 V supply?
I would really appreciate your insight on this point.
Thanks again for your help and advice!
Best regards,
Mingxin Liu
| Stefan Ritt wrote: |
|
We do not have any DC-coupled design for the DRS4. The reason for that is that for many applications we use SiPMs, which are directly powered through the DRS4 board. Since the bias voltage of ~50V would kill the DRS4, we always go with AC-coupling.
The reason the DRS4 evaluation board cannot fully capture signals below -0.5V is the fact that the board only has a single +5V power supply and no +-5V power rails. Having an additional -5V power rail would make the evaluation board too bulky and it could not be powered over USB.
In principle you could bias the point after the AC coupling capacitor to +5V, then you can capture up to -5V signals. Add a divider by two to go up to -10V. But then the input protection diodes won't work as it, you would need some fast 5V zener diodes which open for signals below -5V to protect the following circuitry.
Stefan
|
|
|
|