I've been working with op amp datasheets lately and i noticed strange thing with RF amplifiers specifications. I will provide one example, but there quit more of them with this manufacturer.

HMC625BLP5E http://www.analog.com/media/en/technical-documentation/data-sheets/HMC625B.pdf

So the description says it's "VARIABLE GAIN AMPLIFIER, DC - 5 GHz" But application schematic's show that DC blocking capacitor is required at the input of the amplifier


So my questions are: 1) Why would they say it's DC when it's effectively not. 2) What would happen if i would not put that capacitor on the input?

  • \$\begingroup\$ In RF, anything below 10kHz is "DC". \$\endgroup\$
    – Janka
    Feb 7, 2018 at 7:38
  • \$\begingroup\$ What lower limit do you think they should put on the datasheet? They're selling you a chip, not the inductor and capacitor that will determine the lower frequency limit in your circuit. \$\endgroup\$
    – The Photon
    Feb 7, 2018 at 16:08
  • \$\begingroup\$ As a general rule of thumb, all Hittite datasheets should be taken with a grain of salt :) \$\endgroup\$
    – Selvek
    Sep 11, 2018 at 22:03

3 Answers 3


This isn't an opamp (operational amplifier). It isn't even a differential amplifier. This is a radio frequency attenuator and amplifier on a chip.

  • Why would they say it's DC when it's effectively not.

    Because the inputs of the amplifier and attenuator are DC coupled to the respective outputs.

    enter image description here

    It could theoretically be used for amplifying a signal with a DC component, although in most (probably all) applications you wouldn't want to. The schematic you posted is just a hint, showing a typical application of the chip, it's not the only way to use it.

  • What would happen if i would not put that capacitor on the input?

    It depends wholly on what you connect to the inputs and outputs. To pass the DC component in the first place you'd have to DC couple AMPOUT too (as well as ATTIN and ATTOUT if you wish to use the attenuator at DC, probably those decoupling pins of the attenuator too). Good luck doing that without messing up all the internal biasing of the amplifier, however. I doubt that it could be done, nor that this is the right tool for the job.

I agree that the manufacturers are misleading here. It can't really do actual DC under practical terms, but even 1 MHz is functionally DC when compared to 6 GHz.

  • \$\begingroup\$ Thank you for you answer, it is very helpfull. I'm sorry for the offtopic question, but from you answer i got that you used some simulation tool, could you please tell me what it is? Im currently working on my thesis and it would be very usefull. \$\endgroup\$
    – Joker
    Feb 7, 2018 at 8:13

This is more or less a marketing thing.

As you've noticed, the amplifier (and many many others) need DC-blocks. They are also biased over an inductor at the output. Any DC-signal would just be absorbed into the supply voltage.

The "DC-capable" feature only tells you, that the amplifier itself is not the limiting factor for low frequencies. Many of these RF-amplifiers are basically just a 50 Ohm matched transistor in a common source/emitter circuit. Therefore, the amp is DC capable. Just its standard circuit is not.

Theoretically you could achieve amplification at a very low frequency. You just have to use a very large inductor and huge capacitors for AC coupling. Unfortunately, values become unrealistic at some point. Imagine an inductor that blocks 2 Hz...


They say it is DC if the DC is within a narrow range. For RF only a simple blocking capacitor solves this issue. As an op-amp though with a wide-band response some manufactures use them for Timer/Counter front ends usually with a 10:1 divider.

A servo-loop keeps track of the DC offset by integration so the input always has an average DC value that is correct. Note that not all RF op-amps perform well at DC or even low band AM radio. But they give the spec anyways as a selling point to those who know how to get wide-bandwidth and keep it stable.

Think about oscilloscopes with real-time GHZ inputs that MUST use special RF op-amps that are DC stable-if you use an active servo loop to keep it stable.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.