i like this post (click again to cancel)
1
i dont like this post (click again to cancel) remove favorite mark from this question (click again to restore mark)

Hello everybody. We have an AVANCE 500 MHz bruker instrument. Unfortunately, our amplifier has been weakened and I can not reach 90 degree pulse for X nuclei channel. When I increase the power (decrease the Pl in XWin NMR) my signal always grow. I am going to use long pulses with duration of 80 microsecond to 100 microsecond or even more to reach 90 degree pulse instead of common pulses which are about 10 microsecond. Is this situation hurt any part of instrument? Or I only lose current broad excitation band and will be forced to take several spectrum to sweep expected area?

asked Jul 06 '12 at 08:15

He's gravatar image

He
21


7 Answers:
i like this answer (click again to cancel)
0
i dont like this answer (click again to cancel)

First, I agree with Matthew that the first thing to check is that you're using the "zg" pulse program, not the "zg30". I've run in to this one many times, and so have our students.

Also, I'd like to second the suggestion by "fid" that you should check the peak-to-peak voltage of the signal coming into the amplifier. I'm not so familiar with Bruker RF hardware routing, but I recall on our old Varian Inovas that the RF path went: PTS synthesizer -> Transmitter board (which creates pulses and desired times, but always at a constant power, Vpp = 1 V I recall) -> Attenuator board (which is where the power actually gets set by introducing more or less attenuation in the line) -> Amplifier (which always amplifies the signal by a constant amount). I found on one or two occasions that our Attenuator board had failed, and that gave rise to longer pulses ... and, oh yeah, weird nonlinear dependence of the 360-degree pulse width on poer settings.

Suggestion 2.5: without a scope, you can try determining 360 pulse widths at different powers. When everything's working well, you should get a nice linear dependence when you plot p1 versus pl1^10 (or whatever exponent you need to get the appropriate linearization fo the dB scale). If your amp is fritzing, you may still get a linear dependence. If it's the Transmitter board (which in the Avance I think has the attenuation built into it), you may get wweird nonlinearity, flat dependence, steps, or other strangeness.

You may also wish to check the quality of the RF coming in to the Transmitter board (or simply out if you board has a built-in synthesizer). I've encountered problems before where the PTS output Vpp was low, leading to long pw's, and others where the RF quality was poor (dual sine waves at different amplitudes, noisy signals, etc.)

I prefer to engage these troubleshooting tests before hooking up the scope to the full-power end of the amplifier. It's easy to burn up an oscilloscope if you don't have enough attenuation, and if you don't know that you need one of those big old oil-can 30 dB high-power attenuators you're in for trouble.

One more thing: If you do hook up the scope to the full-power amp, you may also wish to compare the Vpp you observe at the probe versus the Vpp at the amplifier output, thus determining whether the problem is your HPPR preamp unit. Contact Bruker to find out what the expected Vpp drop across the HPPR should be.

But before you do all that, you may wish to measure your 1H and 13C S/N using current calibrations for your 90. If your S/N is way down, you may wish to focus your attention on the probe and preamp. Comparing 1H and 13C S/N with their expected values should help determine whether your problem is restricted to either your high-band or low-band path.

Good luck!

  • Josh
link

answered Jul 19 '12 at 06:24

jkurutz's gravatar image

jkurutz
131

i like this answer (click again to cancel)
0
i dont like this answer (click again to cancel)

If it is an amplifier issue you wish to diagnose, replace the probe with a 50 ohm load, (or a load of resistance equal to that of your probe if it is not a 50 ohm probe) and measure the voltages from the input and output. No sense burning a probe when amplifier diagnostic is better into a simple load than the NMR probe.

If you are permitted, open the probe or amplifier and check for obvious signs of damage. Hot components, burnt components. If you see spikes in the reflected voltage as you adjust a tunable capacitor this can mean you have burnt char on the teflon inside the tunable caps. This is more likely if you do solids work and are putting >1 kW power into the probe. Please be careful as the amplifier can kill a human no problem (use your right hand so high voltage wont pass directly through your heart on the way to ground) and not all labs appreciate one opening NMR probes.

It is very annoying that Bruker does not have the nutation array method of setting flip angles. When I get around to writing my own I will post, if some one beats me to it, please post it.

link

answered Jul 16 '12 at 09:34

w101bdk's gravatar image

w101bdk
7

i like this answer (click again to cancel)
0
i dont like this answer (click again to cancel)

Before you blame the hardware I would suggest checking for simple mistakes , a couple of dumb things that I have done that have confused me about pulse lengths include mistakenly using the zg30 pulse sequence rather than zg to calibrate with and setting the d1 time too short, try 10 or 20 seconds d1

link

answered Jul 13 '12 at 03:24

Matthew%20Revington's gravatar image

Matthew Revington
41

i like this answer (click again to cancel)
1
i dont like this answer (click again to cancel)

Hi Sina,

basicaly u r doing a nutation expt (p1 array) by repeating a 1D expt several times. There may be artefacts there. Auto scaling or things like that may give u wrong impression.

The best is to do a real array expt. I know, it is annoying doing that on bruker. Bruker has no concept for arrays. There are many patches for that (paropt, popt, parray ...) but none of them works properly. They are all very buggy, extremely annoying ...

There is a last one i never tested, called "pulse" or sth like. Or else try popt, is the least buggy among them, i think.

The behaviour of negative peaks being more intense is non sense. The expt is probably wrong. If the long pulses persist at different probes with good tune/match, the only way is checking with an osciloscope since it can be either amplif or console giving too little power. (see 1st post)

cheers...

link

answered Jul 09 '12 at 10:41

fid's gravatar image

fid
139

Hi FID, Thanks a lot for your help. OK. If I could I will try oscilloscope. Best Wishes, - He (Jul 09 '12 at 11:54)

i like this answer (click again to cancel)
1
i dont like this answer (click again to cancel)

Dear Nikolay Thanks for your comment. I am sure about tune and match. I check tree probeheads with a variety of X nuclei (such as 35Cl,37Cl,63Cu,65Cu,199Hg,31P etc ). They all behave alike. I'm not sure that the problem is due to amplifier, it's only my guess. Actually I have been confused. It may be ridiculous but I guess the excitation magnetic field (I mean B1) depends on pulse duration (P1). When I am trying to reach 90 degree pulse I set power level on a specific value. Then I set a value to P1 and take a spectrum, I adjust the phase. After that I increase P1 and repeat above process without readjusting phase. Firstly The peak grows. By setting a good value to P1 the peak vanishes. At this point if I increase P1 I can see the reverse peak. Every things seems Ok till now, but the problem is that the intensity of inverse peaks are very bigger than the positive ones. In fact I can not gain sinusodial curve which expected in this process.

link

answered Jul 08 '12 at 09:12

He's gravatar image

He
21

i like this answer (click again to cancel)
2
i dont like this answer (click again to cancel)

Hi Sina,

Check the match value of your probe, as Nikolay told. You have to reach low wave reflection (>15bB) or else check your probe for defects. If you get a nice tune/match curve with a minimum at your desired frequency and good attenuation, your probe is fine and u dont need to check your expt in another one. Well, check also if the central frequency to which you are tuning is really the right one. For ex if u have a very high ppm scale in your spec.

The best 1st step in such a situation is actually to plug your X-amplifier output through an attenuator (say 30dB) to an oscilloscope and measure the pulse power. Remind that the rms amplitude is =(peak-to-peak ampl)/2.8. Measure the output power at different input intensities and see if you get the specified amount from the manual (around 300W max).

If u do not get the specified output power from your amplifier, measure the input power to it. (here u dont need an attenuator). It must be around 1A peak max, or else the problem is in your console.

Your pulse must be faster than the fastest coupling in your system. In liquids it not a big problem. But then it must be at least faster than the inverse of your spectral width, with a safety margin. For a nucleus at 100MHz with 100ppm spec width its around 10kHz, thus 100us is the red line. Keep away from that.

Long powerful pulses may burn your probe through heating. It may be the case if you need to increase power due to bad tuning/matching.

Cheers...

link

answered Jul 08 '12 at 03:16

fid's gravatar image

fid
139

i like this answer (click again to cancel)
2
i dont like this answer (click again to cancel)

Are you sure, that problem is in your amplifier? But not in probehead or probehead tuning? Sometimes it happens because of loss of Q-factor of probehead (damaged capacitors etc.) - try to check pulses for another probe. Of course, long pulses at high power may damage your probehesd.

link

answered Jul 07 '12 at 15:36

Nikolay%20Boev's gravatar image

Nikolay Boev
41

Your answer
Please start posting your answer anonymously - your answer will be saved within the current session and published after you log in or create a new account. Please try to give a good answer, for discussions, please use comments and please do remember to vote (login to vote)
toggle preview

Tags:

×2

Asked: Jul 06 '12 at 08:15

Seen: 11,192 times

Last updated: Jul 19 '12 at 06:24

powered by CNPROG