Probe Tuning
Probe Tuning Techniques
Why Tune The Probe?
Radio-frequency waves cannot be efficiently sent over regular wires.
To prevent signal losses, "transmission lines" are used. These present a
"characteristic impedance" to, for instance, a radio-frequency generator.
The generator, the transmission line and the load at the other end must
have the same impedance (complex resistance). If the impedances differ,
a "mismatch" condition exists, leading to "standing waves" or "reflection".
This means that the rf power is not efficiently transferred, leading to a loss.
In an NMR experiment, the pulse transmitter is connected to the probes'
detection circuits during the rf pulse. The receiver's preamplifier is
connected to the probe's detection circuit during the detection of the
NMR signal. In addition, the decoupler transmitter, if used, is connected
to the decoupler coil circuits all the time.
With perfect matching, we obtain the shortest possible pulse width,
the best possible signal to noise and, if decoupling is used, the
most efficient decoupling.
Mismatch conditions lead to longer 90 pulses (not a big problem),
loss of signal (thus resulting in lower S/N - definitely a problem)
and inefficient Broad Band decoupling, leading to increased residual
line broadening.
The impedance of the pulse and decoupler transmitters, and the
connecting cables of an NMR spectrometer is usually 50 Ohm . The probe
must therefore present a 50 Ohm load impedance to avoid the above problems.
Basic Probe Circuit
The task an NMR probe circuit is to generate as high a current as possible
from a given input power. For detection of signals, it must generate as much
output power as possible from the fluctuating magnetic field induced by the
spins. This requires the use of "resonant circuits".

The basic NMR probe circuit consists of a coil and two capacitors.
The coil and the capacitors, C1 and C2, form a resonant circuit. The
inductance of the coil, L, and the (approximate) sum of C1 + C2
determine the resonance frequency. C2 provides transformation of
the high impedance parallel resonant circuit to a lower impedance
(generally 50 ohms) to match the output impedance of a transmitter
and the input impedance of a receiver.
Probe Tuning
If the probe is not properly tuned and/or matched, rf power is not
optimally transferred, thus lengthening the 90° pulse width
(for a given transmitter output) and worsening signal to noise.
In the case of a decoupler coil, the decoupler efficiency is reduced.
Tuning a probe is obviously necessary when a different nucleus is
to be observed with a so-called broad-band probe. Not so obviously,
it is often necessary after changing the sample! This is due to
the sample's rf-properties. Its dielectric constant affects the
probe tuning by (simplistically speaking) adding a capacitance
to C2. The sample's conductivity results in "loading" the circuit,
the effect is similar to placing a resistor across the coil.
This means that the impedance transformation is affected.
C2 must be increased, and C1 must be decreased.
To tune a probe, we need a frequency source, a display device and
some form of "impedance sensor." Two simple "sensors" are
generally used, an impedance bridge or a directional coupler.
The output can be displayed on a meter or an rf-oscilloscope.
Tuning For Different Samples
The type and size of the probe in use often determines the need
for tuning after changing samples. Generally, the higher the
frequency, and the larger the sample size, the more the sample
affects the probe tuning. The sample's electrical properties are
important as well; highly ionic samples (buffered H2O or D2O
solutions) cause much more de-tuning than a sample in CDCl3.
The arrangement of the probe coils matters as well. The
proximity of the sample to the coil is an important factor -
the closer the sample is to the coil, the more it will affect
the tuning. A probe configured for direct detection of
X-nuclei has the X(low)-frequency coil close to the sample
and the proton (higher frequency) coil on the outside. Most
samples will not noticeably change the proton tuning or the
carbon tuning (except perhaps for aqueous solutions as noted
above). With an Inverse Detection probe, the coil arrangement
is reversed and different samples will more likely change
the probe's proton tuning.
For example, if a regular probe is tuned to 20 MHz, changing
a 5 mm sample will hardly change the probe's 20 MHz tuning.
With an Inverse Detection probe at 600 MHz, using 10 mm samples,
the proton circuit will most certainly need tuning for each sample.
Using the Bruker Reflection Bridge
Bruker provides a reflectance bridge, a combination of an
impedance bridge and metering circuit. While rapidly pulsing
(less than 50 ms repetition rate), with pulses 20 to 100µs
long, the reflected power is observed. The amount of reflection
is a measure of the deviation from the expected 50 W impedance.
The meter range is adjustable by factors of two.
The original Bruker unit is meant to replace the preamplifier
in the preamplifier housing while tuning the probe. We have
modified the units to avoid wearing out the fragile connectors
by frequent swapping of the bridge and preamplifiers. To connect
the modified units, remove the BNC cable at the appropriate probe
connector 1H for proton, X for other nuclei) and plug it into the
matching cable on the bridge. Connect the other cable of the bridge
to the probe.
With X-nuclei, a reading of 1 on the x 5 range is probably the
lowest one can expect. For protons (and fluorine), it should be
possible to tune for a reading of 1 on the x 10 range.
Tuning Procedure
If the probe is tuned while it is in the magnet, the sample
should not be spinning to give a stable reading. Most contemporary
probes contain an observe and (at least one) decoupling circuit.
The tuning of the circuits usually interact, meaning that if one
of the circuits is being tuned, another might become de-tuned.
Higher frequency circuits are affected more than lower frequency
ones. It is therefore important that the highest frequency circuit
(normally protons) be tuned last. If the X-frequency circuit is
to be tuned, the decoupler should be turned off to prevent
false readings. If a circuit needs to be tuned for a particular
sample it is often sufficient to only re-adjust C1, the "Tune"
capacitor. The dielectric properties (highly ionic solvent) of
the sample may require that C2, the "Match" capacitor, be
readjusted as well. As the Tune is changed, re-adjust the
Match for minimum reflectance. Repeat this procedure until a
minimum is reached.
Once the ratio of the maximum reading with the probe
de-tuned and the reading with the probe tuned is better
than 20:1, no greater performance improvement will be
obtained from further tuning.
Tuning Probes on the AM and AMX Spectrometers

The Broad Band probes have rotary variable capacitors to tune
the proton circuit and sliding "digital" capacitors for the
X-frequency circuit.
The H-C Inverse probe (AM-500) and the Dual C-H probe
(AMX-300) use rotary capacitors for proton and carbon.
The QNP probes are more complex: Users must not attempt
to tune carbon etc., if tuning is needed, let the facility
staff take care of it! At most, proton tuning might be advisable!

AM-400/500
1H
For proton tuning, the decoupler can be used, BUT: first, the range must be switched so the system is set up for X-nuclei observe! Otherwise components in the pre-amp housing will be destroyed!
1) Turn the spinner off. Type 'PR H 2' followed by II (or,
easier, RJ CARBONx.x00, II) then DP = 8H (decoupler level in
high power mode). Type 'CW' to turn it on. Set Console switch
to DEC REFL to display reflected decoupler power.
2) Minimize reflected power with the YELLOW Tune control,
perhaps together with the Match control.
3) Type 'PO' to turn decoupler off. If a parameter file
exists, read in the proper parameters - otherwise, type
'PR' = 1 1. Type 'II' to set the interface to the new PR
parameters.
Alternatively, the "tuning brick" (reflection bridge) can be used, see below.
X-Nuclei:
1) To tune the observe coil, use RJ with the appropriate
parameter file. Install the reflectance bridge between probe
connector marked "BB" or "C" and the cable normally in place.
With the Broad Band probes, set the controls Tune and Match to
previously established numbers. Type 'PRBT' (or 'EXE PROBETUN')
to start rapid pulsing. Minimize meter reading with the Tune and
Match controls. Switch the meter range as needed. At the "5"
position, a reading of 1 is about the best you can hope for.
A reading of >3 in the "4" position is marginal. Press CTRL/H
when done. Remove the reflectance bridge and re-connect the
regular probe cable.
2) Tune the proton coil as described above.
It is also possible to switch the bar display to OBS REFL.
No reliable tuning is likely though!
AMX-300/400:
The reflection bridge is normally used for tuning. Select
a parameter set appropriate for the nucleus. Type (xau)
"PROBETUNE" to start the rapid pulsing. Tune the probe as
described above. Click on Task -> Kill, select "go" to
stop pulsing or press CTRL/U. The previously used parameters
are restored after the xau program has finished.
Bruker Console Indicators
Bruker NMR spectrometers have a directional coupler permanently
installed as part of the rf power amplifiers for the observe
and decoupling channels. Only the AMX-series spectrometers
are capable of simultaneous display of output and reflected power.
The pulse power amplifier's reflected power indicator should
only be used when there is no external attenuator connected
between the output and the probe (as might be required to
prevent arcing of probe circuits as a result of too high an
rf power). The attenuator also reduces the reflected power,
thus providing a low reading even if the probe is not well
tuned! In such a case, the Bruker reflectance bridge should be used.
The decoupler contains a directional coupler as well. Using
the decoupler for (proton) tuning, the reflected power is
minimized by adjusting C1 and C2.
Changing Probes, Tuning for Other Nuclei
Save the current (hopefully useful) shim settings to a temporary shim
file before removing the regular probe.
After Restoring Normal Probe and/or Tuning
After reinstalling the regular probe, read the temporary shims
back in and shim the X, Y, XY, XZ, X2-Y2, ZX2, YX2 without
spinning, and the Z1-Z3 gradients with spinning. Store the new
settings in SHIM[mmdd] where [mmdd] is e.g. 1005 for October 5th.
After a probe change or tuning the performance of the regular
probe must be checked on protons and carbon (as well as 31P
and 19F on QNP probes).
Relaxation times for benzene-d6 are very long! Wait at
least 300 seconds between pulses if you want to check S/N again!
If you change or re-tune probes, it your responsibility
to assure that the system is back to normal after you
are finished. The regular standard sample must be in the
probe on all instruments. It must locked, spinning, and
auto-shimming on Z1-Z3.
AMX-300
The Inverse Broad Band probe is normally used. It must be
left tuned for 13C.
VBAMX-400, AM-400
The QNP probe normally used is permanently tuned for 1H, 13C,
31P and 19F. For normal operation it should not need tuning.
The QNP preamp can be used with the BB probe. Users need to
know how to change not only the probes but the pre-amp boxes
and the pneumatic switching actuator as well. Aligning the
switching actuator, checking the QNP probe performance and
tuning is quite different (for instance, tuning must always
be started on 31P) from the regular probes. It often requires using a directional coupler and an oscilloscope.
To assure proper operation, facility staff must be consulted
before changing probes. Users may, on occasion, be permitted
to install a different probe but the facility staff must
perform the testing and tuning (if required) after the QNP
probe is re-installed, Please allow for up to three hours
when reserving instrument time!
AM-500
The 5 mm Inverse H/C probe must be installed and tuned for 13C after you are done.
AMX-300
After re-tuning the probe, determine the 90 pulse width for 13C
and use the xau program "astmtest". S/N should be > 50:1.
For verifying the proton performance, determine the 90 pulse
width and, with the 0.1% ethylbenzene sample, run use the
xau program "protsn". S/N should be > 150:1.
Note 90 pulse widths and S/N in the log book and give us
the resulting plots.
VBAMX-400
After re-tuning the probe, determine the 90 pulse width for 13C
and use the xau program "qnptest". S/N should be > 100:1 for
13C and 31P, at least 80 :1 for 19F.
For verifying the proton performance, determine the 90 pulse
width and, with the 0.1% ethylbenzene sample, run use the
xau program "protsn". S/N should be > 130:1.
Note 90 pulse widths and S/N in the log book and give us the resulting plots.
AM-400
With the ASTM sample, use 'EXE QNPTEST' to check 13C and
31P performance. S/N on both should be > 100:1
To verify proton performance, measure the 90 pulse width,
with the 0.1% ethylbenzene sample, run 'EXE PROTSN.400' .
This EXE will run a single scan and plot the result. S/N
should be > 130:1.
Give us the plots from the QNPTEST and this test. We
can quickly tell if something is amiss. Note the 90
pulse widths and the S/N in the log book.
AM-500
With the ASTM test sample, use 'EXE ASTMINV=D1' to
check the carbon Signal/Noise ratio. The S/N of the
C6D6 triplet should be better than 75:1. (The coupled
dioxane triplet should show better than 40:1.) Note the
S/N in the log book. Plot the spectrum.
To verify proton performance, measure the 90 pulse
width, with the 0.1% ethylbenzene sample, run 'EXE
PROTSN.500' . This EXE will run a single scan and
plot the result. S/N should be > 320:1.
Give us the plots from the ASTM and this test.
We can quickly tell if something is amiss.
Other Tuning Methods
Other methods are used mostly away from the spectrometer.
If a probe cannot be tuned on the spectrometer, it may be
so severely de-tuned that finding the proper adjustments
is nearly impossible, or it might be defective. Several
measurement arrangements can be used, depending on the
availability of equipment.
An impedance bridge or a directional coupler can conveniently
be used for probe tuning. A frequency synthesizer (or stable
generator) is needed for an rf source or the spectrometer
can be used to supply either pulses or low power CW signals.
An (rf) oscilloscope is usually needed to display the output.
One of the simplest methods is the use of a sweep generator
(a further option, the bi-directional Watt-meter, will be discussed later).
Using a Sweep Generator
A very convenient set-up is to use a sweep generator to supply
an rf source which rapidly changes the frequency through a
selectable range. The output of the impedance bridge or the
directional coupler can be rectified with a diode detector.
The sweep generators usually have an X -output which can be
used to drive the X-axis of an oscilloscope (which must be
set to X-Y display mode). Sweep generators usually have
frequency markers which are used to display a blip on an
oscilloscope for calibration of the X-axis. This feature
is often only usable together with an internal diode detector.
In this case, the sweep generator also has a Y-output.
The oscilloscope displays reflectance versus frequency.
(In this case, the oscilloscope need not be an expensive rf-type.)
The use of a sweep generator is very convenient for tuning probes
where the resonance may be far away from where it should be.
Impedance Bridge

These devices can be very broad band but above 200 MHz, some
units may be of limited usefulness.
Directional Coupler

The directional coupler has rf-properties which make it useful for
tuning circuits. It has a low-loss path between IN and
OUT.
A large loss (or small coupling) occurs between the COUPLED
port and the OUT port. A smaller loss (or closer coupling)
occurs between the COUPLED port and the IN port. The loss
is expressed in decibels (dB).
Typical properties are:
IN - OUT: -0.1 dB
OUT - COUPLED: -40 dB
IN - COUPLED -10 to 20 dB
If an rf-source is applied to OUT port, practically
the full level will appear at IN port. A small signal
(-40 dB) will appear at the COUPLED port. If the
IN port is terminated into 50 W , no power will be
reflected. If mismatched, the reflection will not only return
to the OUT port but also, at 10 or -20 dB, at the
COUPLED port. Thus, if a scope is connected, we can
observe the reflection. Directional couplers can be very
broad band. If in doubt, consult manufacturers specs.
Using an In-Line Bi-directional RF-Watt Meter
An in-line rf Watt meter can be used together with a
spectrometer's amplifiers. Normally, only the decoupler is
used; the observe power amplifier output is often too high
to be turned on for any length of time without burning out
the probe circuits, or it may not be able to operate in
continuos wave mode (CW).
The meters use a one (or two) plug-in(s) of appropriate
frequency and power range. Rotating the plug-in selects
forward or reflected power display.
The biggest advantage is simplicity of operation.
The drawbacks include the need for different plug-ins.
Only the more elaborate models are capable of operating in
pulsed mode. This means that the simple variety cannot be
used on many of the Bruker spectrometers for tuning nuclei
other than Protons.
Questions, comments to: Rudi Nunlist
rnunlist@bloch.cchem.berkeley.edu /tt>
Last Update: 4/18/95.