Could some electronics guru please clarify the following a bit ? Any schematic for a Gilbert cell (multiplier) has inputs labelled as LO+, LO- and RF+, RF-. Normally, signals are referenced with respect to ground, so what are the + and - for ? Any hints would be useful.
Those are differential (balanced) inputs. I'd have to see a data sheet to see if it's OK to ground one of the inputs, making it single-ended.
Good Luck! Rich
The common mode voltages have to be within certain limits (top higher than bottom, bottom higher than V-), but the CMRR is generally high due to the differential structure. So as long as you satisfy the ground referenced limits, the ground referenced voltage has little effect, and the signals are referenced to each other, i.e. (V+ - V-), rather than (Vsig - Vgnd).
Tim
-- Deep Friar: a very philosophical monk. Website:
Uhh.... for what frequency? That might work at DC but you'll find surprises in the high frequency limit, in terms of phase shift and amplitude.
Since common mode "doesn't matter", you can bypass (not ground, since it needs the correct common mode offset) one of each input and use it single ended.
This becomes more complex at HF, since capacitances become an issue. Balance helps to suppress carrier. That said, I've seen SiGe Gilbert cells that claim single ended operation out to UHF.
Tim
-- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms
And remember that there's DC ground and AC ground. Sometimes you can tie half a differential signal straight to circuit ground. Sometimes you can connect to ground with just a bypass capacitor. Sometimes you need a resistor/cap, for bias current. Sometimes you need a resistor or a coil to the opposite input. Sometimes the circuit is sensitive to common-mode signals, and you absolutely positively must provide it with a balanced signal.
Read the data sheet. It doesn't tell all, but it certainly tells lots. Often you'll find a recommended circuit, or a test circuit. If you have specific questions about a specific part _after_ you've read the data sheet, ask -- we'll help.
-- Tim Wescott Wescott Design Services http://www.wescottdesign.com Do you need to implement control loops in software? "Applied Control Theory for Embedded Systems" was written for you. See details at http://www.wescottdesign.com/actfes/actfes.html
I just read of your site that you got a job dealing with induction heating. Congrats and good luck
A Gilbert cell is a differential device; unless you have some assurance that grounding will correctly DC bias the port of interest, you cannot make that port arbitrarily a ground-referenced input.
Unless the IC has internal biasing and level-shifting, you must bias the cell yourself. Access to the differential pins is required. Often, the output pins are differential, as well.
Cell phone base stations almost certainly do all the baseband processing in digital-land, then output an I and Q channel via DACs. At that point they'll be mixed up to RF by either a Gilbert cell or a diode-ring mixer, depending on the prejudice of the designing engineer.
I strongly suspect that a cell phone handset does the same thing in reverse, except that the downconversion is going to get done inside a chip, and will almost certainly be a Gilbert cell. The dynamic range of a Gilbert cell is inferior to a diode ring (or FET ring) mixer, but the power and money savings would overwhelm that consideration in a handset.
-- http://www.wescottdesign.com
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.