PARAMETER	REMARKS
Supply	+ 5 V, 250 mA
Input	1000.1 - 1999.9 MHz
Output	0.1 ... 999.9 MHz

Test Circuit, drawing coutesy of Analog Devices

The datasheet says "The LTC5510 is a high linearity mixer optimised for applications requiring very wide input bandwidth, low distortion, and low LO leakage. The chip includes a double-balanced active mixer with an input buffer and a high speed LO amplifier. The input is optimised for use with 1:1 transmissionline baluns, allowing very wideband impedance matching. The mixer can be used for both up- and down-conversion and can be used in wideband systems." here, we use it for downconversion. The LO used is a XLP736A00.000000I (The Renesas XL devices - XO and VCXO options - are ultra-precision crystal oscillators with 750 to 890 fs typical phase jitter over 12 kHz to 20 MHz bandwidth.) running at 1000 MHz.

Pin 8 - Bias Adjust Pin. This pin allows adjustment of the internal mixer current by adding an external pulldown resistor. The typical DC voltage on this pin is 1.8 V. If not used, this pin must be left floating. As we use a Vcc of 5.0 V and the output range is below 1.2 GHz, according to Table 7 in the datasheet, the recommended Values for R1 is "open".

The assembled PCB will cost approximately € 25.- (2026)

Component	Price [€]	Distributor
SK56C	0.285	Mouser
XLP736A00.000000I	3.74	Mouser
LTC5510IUF#PBF	10.89	Mouser
LM317EMP	1.42	Mouser
TC1-1-13M+	3.29	Mouser
MABA-008639-TC41T7	1.42	Mouser
12 x Capacitor à 0.20	2.40	...
7 x Resistor à 0.10	0.70	...

The only thing to mention here is, that this is a LVPECL thing. This means Low-Voltage Positive Emitter-Coupled Logic. So we have to use resistors at the output (Emitter) to create a voltage. (typically 700 mV to 800 mV peak-to-peak). As we use ac coupling and don't need to meet logic level restrictions, we can get away with just two resistors connected to ground - we have chosen 2 x 150 Ω (resulting in about 10 mA of Bias Current) and of course two capacitors (100 pF each). The input impedance of the LTC5510 (LO+ and LO- each) is internally matched to 50 Ω so this is clearly not matched for maximum Power. But the traces are very short with respect to the wavelength, so we don't care that much.

LO Feedthru, as seen at the output

The webmaster does not read these comments regularely. Urgent questions should be send via email. Ads or links to completely uncorrelated things will be removed.