Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

reflectionless impedance bridging

Status
Not open for further replies.

crossflow

Newbie
Joined
Jun 19, 2017
Messages
4
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
66
let's suppose we have a fixed low-impedance source and a fixed high-impedance load with the purpose of maximum voltage transfer, to be connected each others by a long transmission line compared to wavelength of signal transmitted and we want minimal distortion caused by reflections. Is it a trade-off or is there a solution to have both maximum voltage transfer and a matched reflectionless line?

When referring to the impedance bridge connection, we usually think of audio applications, obviously in that case there are no problems. I refer to those applications with ultrasonic transducers between HF and VHF and meters of cable, drived with sine waves or square waves of MHz and hundreds of volts. For example inspection probes, non-destructive testing or others types of imaging.
 
Last edited:

A low-reflection setup requires cable impedance matching at least at one end. Source side series termination is the most common method.

It's unlikely that the said ultrasonic transducers can be considered as high impedance load.
 

Just to clarify, it's not absolute high impedance but at resonance frequency minimum impedance of transducer is ten to a hundred times the source impedance, source impedance (Rdson) is low to maximize voltage transfer to the load (pzt require high voltage) and to charge the parallel c0 capacitance of transducer as fast as possible, adding series resistor limits the output current and voltage even at DC and it's not acceptable in efficiency critical designs. Rlc load is quite complex and i don't know how to adapt correctly for a wide range of waveforms and frequencies.
Anyway, my question was about the general problem when we have to provide impedance matching/bridging in an application that require both:
trade-off or not?
 

it's not absolute high impedance but at resonance frequency minimum impedance of transducer is ten to a hundred times the source impedance...(Rdson)

But not very high compared to characteristic cable impedance. Taking Rdson as source impedance sounds plausible at first sight, until you hit the keyword "VHF".

adding series resistor limits the output current and voltage even at DC and it's not acceptable in efficiency critical designs
That's another word for accepting cable reflections instead. A resistive termination is also the key for maximum transducer bandwidth.

The alternative approach is to use a fixed cable length with the transducer and treat it as part of the transducer impedance.

It should be mentioned that dedicated wideband ultrasonic transducers are containing lossy elements themselves, e.g. an absorptive backing of the piezo disc.

adding series resistor limits the output current and voltage even at DC
Hardly when driving a piezo transducer load with capacitive impedance below the first resonance.

Do you have a specific transducer with known impedance characteristic in mind?
 

But not very high compared to characteristic cable impedance. Taking Rdson as source impedance sounds plausible at first sight, until you hit the keyword "VHF".
Z0 of connections (pcb traces and cable) is one of those things that can be arbitrarily varied within certain limits, while source and load impedance are fixed. Taking Rdson is correct if the silicon device productor declare VHF square wave operation and specifies minimum dv/dt output at various load conditions.
In a flipped wafer level chip, there is no bonding, parasitic inductance or capacitance are neglible.
That's another word for accepting cable reflections instead. A resistive termination is also the key for maximum transducer bandwidth.

The alternative approach is to use a fixed cable length with the transducer and treat it as part of the transducer impedance.

It should be mentioned that dedicated wideband ultrasonic transducers are containing lossy elements themselves, e.g. an absorptive backing of the piezo disc.
in worst cases reflections could damage devices or reduce the bandwidth performance, reduce the efficiency or distort waveform creating cabling dependent images. Fixed cable also must be characterized before use.

Hardly when driving a piezo transducer load with capacitive impedance below the first resonance.

Do you have a specific transducer with known impedance characteristic in mind?
the source drives the transducer at the resonance frequency, typical real part load value is 100-1k, parallel capacitor is about 100p-1nF.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top