LHM6554LE configuration

Hi Niklas,

For amplifying the very low level signals you have shown, I would not use the LMH6554 which is a current feedback amplifier which has restrictions on the value of its feedback resistor, etc. A big question to answer is what is the frequency range / highest frequency of interest and what kind of SNR are you trying to get?

Fro now, I will draw your attention to some good discussion / information on getting a high gain amplifier to work, here:

http://e2e.ti.com/blogs_/b/analogwire/archive/2013/01/25/high-gain-high-bandwidth-how-can-i-get-it-all.aspx

http://e2e.ti.com/blogs_/b/analogwire/archive/2013/02/18/high-gain-high-bandwidth-putting-it-all-together.aspx

http://e2e.ti.com/blogs_/b/analogwire/archive/2013/02/01/xavier-ramus.aspx

Please review the Analog Wire links above and let's discuss afterwards if you need any help.

Regards,

Hooman

Hi Hooman,

Thank you for the fast response and the links. The highest frequency of interest is in the range of 10 MHz. Unfortunately I will not be able to etch a board by myself. Since it is possible to use a second amplification stage, the gain is not so important for my porpoise.

As a differential input/single ended output amplifier, the LMH6629 is suggested in the links above, which is available with an evaluation board. Here the impedance mismatch could be a problem. Our samples have a resistance in the range of 10 kΩ. So the input impedance should be in the range of 1 MΩ. Do you have any suggestion for usage of this amplifier?

Best wishes,

Niklas

Hi Niklas,

So, since you cannot do your own board,  I think you are looking for an Op Amp which fits the requirements and has an EVAL (EVM) available. With your large source resistance (10kohm), I'd think you would need to use a high gain, CMOS or JFET input (for high input impedance) amplifier. Here is a list of high speed (>200MHz) that fits this:

OPA653

OPA656

OPA657 (low noise)

THS4601

LMH6601

OPA2354

OPA2355

OPA2356

OPA2357

OPA3355 (low noise)

OPA354

OPA355 (low noise)

You would also want to look through this list and select the ones which have the lowest 1/f noise corner (to minimize the noise impact).

Of the list above, I think the OPA657 holds the best promise because of its lower noise, with 1/f noise corner of about 2kHz,  FET input, and high (1.6GHz) bandwidth which should allow you to operate at a high closed loop gain and not be band-limited. It does have an unpopulated EVM available also:

http://www.ti.com/tool/dem-opa-so-1a

Regards,

Hooman

Hi Hooman,

Thank you very much for your help. I will try to use the OPA657 amplifier for the reasons you mentioned above. Also the evaluation board http://www.ti.com/tool/dem-opa-sot-1a looks suitable, since it can be used to build a differential input/single output amplifier. Attached I put a plan, showing how I will equip the board. Is that possible? Also I tried SPICE simulations. How reliable are such simulations?

There is one other question: The manual of the evaluation board suggests to use a tantalum capacitor with 2.2µF, 16V and size 3548. Unfortunately I cannot find a capacitor with seize 3548. Can I use another capacitor size instead or can you suggest where to look for such a capacitor?

Best wishes,

Niklas

Hello Niklas,

The circuit you have attached won't have much gain (low R_F, and large R_G!). Besides, if your intention was to have a differential gain, the gain from your source, bottom-side to the output is almost 0 while you have a gain of 11V/V from the top side of your input! So, I'm not sure if this is the circuit you are looking for because it is so imbalanced.

My recommendation would be to either do your own low noise, high impedance instrumentation amplifier, or get one off-the-shelf.

Here is what I mean by instrumentation amplifier (high and equal input impedance and equal and opposite gain from each input) (from AN-31):

Or you find an off-the-shelf one that fits your noise, bandwidth, input impedance, power consumption, input voltage range:

http://www.ti.com/paramsearch/docs/parametricsearch.tsp?family=analog&familyId=500&uiTemplateId=NODE_STRY_PGE_T

With regards to Pspice, most TI macromodels include a variety of parameters modeled. For example, in the case of the LMH6554, here is what the device model "comment" lines state as the limitations / inclusions of the model:

* Legal Notice:  
* The model may be copied, and distributed without any modifications;
* however, reselling or licensing the material is illegal.
* We reserve the right to make changes to the model without prior notice.
* Pspice Models are provided "AS IS, WITH NO WARRANTY OF ANY KIND"
*////////////////////////////////////////////////////////////////////
*
*  Model features include: (typical values)
*     closed loop frequency response vs gain
*     output current and voltage limiting
*     input bias current error (common mode)
*     input bias current error (differential mode)
*     input common mode voltage offset
*     slew rate and step response
*     vnoise, referred to the input
*     inoise, referred to the input
*     output balance error vs frequency
*     common mode frequency response
*     harmonic distortion
*     Isupply temco
*     Isupply dependence on supply voltage
*
*  This model does not capture the enable/disable function

For decoupling capacitors, the value or size is not the most important thing about them. They should have low inductance paths to ground and to Vcc and they should be rated for the frequency of operation (below their resonance frequency). Tantalum capacitors are usually good for low frequency to act as charge storage for load transients.

Regards,

Hooman