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.
Hi Supratim,
The reason the JFET LNA design is shown in the LMH6518 datasheet is to maintain the 1Mohm || 5pF input impedance that most passive 10x oscilloscope probes are designed for. If you don't need to interface to an oscilloscope probe of this kind, you don't need to use the JFET LNA, provided you strive for the following, as stated next to Figure 62:
Hi Hooman,
Thanks for your prompt reply. If I use a FET input wide band OPAMP like OPA656 (I/P impedance 10^12 || 1pF) and terminate the non-inverting I/P with 1M || 4 pF, will it be able to replace the LNA? Of course I have to set a common mode of 2.5V. I want to use the same attenuators as shown in the datasheet. Replacing the discrete component LNA by an IC will reduce the component count considerably. The condition of low drift may not be met. But otherwise will it be a feasible solution? Please advise.
Regards,
Supratim
Hi Supratim,
I cannot picture the schematic you have in mind of the OPA656 interfacing your probe signal (ground referenced) and then producing an output for the LMH6518 which is at a fixed 2.5V CM level. You will probably have to use the OPA656 in an inverting configuration which negates the advantage of utilizing its ultra high input impedance. May be you can send me the schematic you have in mind so that we can discuss in more detail?
You can consider the use of the OPA656 as the input interface to the scope probe. However, please keep the following in mind:
1. Bandwidth is lower (500MHz) compared to the LMH6518 (900MHz). OPA656 Slew Rate (290V/us) could be another limitation with LMH6518 at lower gains (higher input swing) conditions
2. Noise will be impacted (7nV/RtHz) vs. 0.98nV/RtHz for LMH6518
3. When doing the 2.5V CM shift with this stage, you end up using it at higher Noise Gains (which means even lower bandwidth).
Regards,
Hooman
Hi Hooman,
I will attach a schematic. But before that I would like to explain my idea more clearly.
I want to use OPA656 as a unity gain non-inverting amplifier (buffer) just after the attenuator stage.
Next stage will be a wide band amplifier which need not have a very high input impedance. This will provide 2.5V common mode.
I completely agree that the noise performance will degrade and I am targeting only 250 MHz BW.
Please suggest me some device which can protect LMH6518 input from over drive and at the same time does not degrade signal quality (very low capacitance). I tried with diode clamp (BAW62), but could not achieve a good clamping.
Regards,
Supratim
Hi Supratim,
Regarding protection of the LMH6518 inputs from overdrive: What was the issue you encountered with using the BAW62 diode for protection? Was it slowing down your bandwidth too much or something else?
If capacitance is an issue, here are some low capacitance schottky diodes to consider as well:
http://www.onsemi.com/PowerSolutions/product.do?id=MMBD101L (7V max reverse, 1pF)
http://www.skyworksinc.com/Product.aspx?ProductID=494 (8V max reverse, 0.83pF)
Regards,
Hooman
Hi Hooman,
I connected the BAW62 diodes at the input of the JFET LNA (i.e. after the attenuator stage). They were reverse biased by 0.8V. The bias voltage was generated by a potentiometer followed by an OPAMP buffer (LMV842).
They were loading (distorting) the input signal slightly even at signal voltages much below the clamping voltage. The clamped voltage kept on increasing with input signal amplitude and the clamped signal was a highly distorted version of the input beyond 10MHz. The lead inductance may play a role in the high frequency distortion.
Thanks for the suggestions. In an oscilloscope application, the input voltage may go well beyond 80V when the front end attenuator may be set at 1X. So the over drive protection should be at least for 80V. Skywork's SMS3924 has 70V max. reverse, 1.5pF. But higher reverse voltage comes with a higher VF. It has VF of the order of 1V. Is there any other way of doing it?
Regards,
Supratim
Hi Supratim,
I've not worked with protection circuits much. So, you may need to double check information I provide.
Skyworks Limiter diodes are another option:
http://www.skyworksinc.com/Product.aspx?ProductID=105
http://www.skyworksinc.com/uploads/documents/BRO372_11B.pdf
I've also been told by colleagues that "Littlefuse" manufactures TVS (Transient Voltage Suppressors) which may also work here (but these tend to be higher capacitance and I don't know what to expect when these are subjected to steady state overvoltage, as in your application?):
http://www.littelfuse.com/products/tvs-diode-arrays.aspx
http://www.littelfuse.com/products/tvs-diode-arrays/general-purpose-esd-protection/sp725.aspx
Sorry if I'm not of much help.
Regards,
Hooman