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

Hey folks, I was wondering if anyone else has had any issues with the experiments/pulse sequences that came standard in the VNMRJ software? I am running VNMRJ 2.2D on Red Hat Enterprise, but have yet to have much success with any experiment outside of standard 1H, 13C, 19F, and 31P. Lately we have been trying to run C13DEPT, C13APT, (HC)HMBC, (HC)HSQC, and HETCOR, and have come up with nothing usable. The DEPT experiment produces the 4 expected spectra, but three are blank/noise and the fourth is just a standard carbon scan; the APT was promising, but I noticed a peculiarity about it: if I printed out the c13 spectra, the peaks were positive/negative as expected until about halfway through, where the spectra repeated itself, only upside down and in the opposite direction (as if it were simply rotated 180 degrees around the center of the spectra); HMBC, HSQC, and HETCOR all ran, but when I try to open them, I only get the message "LP analysis failed." Any attempt to display either fid or the 2D comes up fruitless (after wft each fid independently, selecting a 2D display returns the message "No 2D data in data file"). I have been able to open the HETCOR in a separate software suite (ACD labs Spectrus Processor), and the spectra there are horrific: the proton spectra is a poorly defined mess of peaks mirror imaged around the tof, and the carbon scan comes up as simply noise.

* UPDATE * Since I can only comment 300 characters, I decided to add my response/update as an edited addition.

To Uli: Sample volume and depth are spot-on, aligned with the field. I have tried both manually shimming and gradient shimming, in addition to gradient shimming and then manually fine-tuning the coils. Our nuclei tuning are both done manually. All of the 1D's came up beautifully, and I have run a 13C before each DEPT and APT, and a 1H and 13C before each 2D experiment. I know wft is only for 1D, and I only used it to observe the single spectra by themselves to see what the NMR produced in 1D. Unfortunately, I seem to be lacking this magical green "auto processing" button for 2D, but have been able to process them without it. So, I can safely say shimming and tuning are not the problem :)

Interestingly enough, I changed the decoupling parameters for a HETCOR experiment after noticing that, though the text boxes showed the pulse width and power for decoupling, entering "pp?" and "pplvl?" into the command line yielded '0' for both. I manually entered in the values, and subsequently got a 2D HETCOR; the only problem is, the carbon channel shows a mirror image of itself, (just as in the APT) making the data essentially useless. I'll try to get an image of it. I also ran an array of pw for a standard 1H scan just to ensure that the power hadn't degraded, and that came up exactly as expected, so no problem there.

Thanks for the help!

asked Oct 31 '12 at 03:51

CaymanDave's gravatar image


updated Nov 02 '12 at 06:03

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


First you have to prepare your sample properly! Is the volume in the sample correct and also the depth?

Do you shimming manually or do you use gradshim first?

Do you use ProTune for tuning and matching or do you have done it manually, both channels (proton and also carbon)?

Then do first the 1D experiments to check, if these are fine. Then you can start to do the 2D, HSQC, HMQC and HMBC are basing upon the proton 1D experiment before.

DEPT, APT and HETCOR you must have before a proper carbon 1D to do these experiments afterwards. These base on a the carbon 1D experiment. This is the measured dimension.

If the 1D experiment, Proton and Carbon, are bad, ... then you have not to do the 2Ds any more!

The message about the Linear Predicition is fine, ... there are data points missing to do a correct predicition. You can click on the linear predicition button, to reset the parameters based on the already measured ones, may be it is now working fine.

wft is only for processing 1D's! wft2da or wft2d is the command for 2D processing, please take the green button "auto processing" in the processing menue for the processing stuff.

Do not forget Shimming and Tuning/Matching before the measurement!

Yours sincerely



answered Nov 02 '12 at 03:25

Ulrich%20Haunz's gravatar image

Ulrich Haunz

updated Nov 02 '12 at 03:28

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

OK, there are some possiblities left!

Do you can send or implement a picture of the artefact? There are different types of artefact, you can see, if something goes wrong! Do you see the artefact in VnmrJ 22D? Or with ACD-Labs software?

Is it the centre spike - direct in the middle of the spectrum? Or do you see tall peaks minimized and inverted on the other side?

Have a look at the following page/link:


There are also some notes to the artefacts, maybe this helps you.




answered Nov 02 '12 at 07:33

Ulrich%20Haunz's gravatar image

Ulrich Haunz

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

We have observed something similar at times because of probefile mismatch. Make sure your probe file is set the right one, as at times a wrong probe file might give good proton 1Ds but might have problems with 2Ds. The probe file can be accessed from the radio button "Probe" next to the spectrometer status box (Idle/Active). Also, make sure your chempack version is up to date. That will take care of any pulse sequence and parameter issues.

About the "LP analysis failed", I think it is because of the linear prediction parameters have not been set fine. The best thing to do to start with is to turn off LP from the processing window and see what you get.

The "f1coeff" parameter is critical for processing 2Ds. For experiments that involve normal quadrature detection in the indirection dimension, use f1coef = ''. Use appropriate ones for sensitivity enhanced experiments.

Also, what is the parameter fsq (frequency sensitive quadrature) set to?

These are my 2 cents, hope they are of help.


answered Nov 02 '12 at 09:46

Bharathwaj's gravatar image


Probefile and chempack are properly set and up to date. I asked the value of the f1coeff parameter ("f1coeff?") and it responded with "Variable doesn't exist." fsq is set to 'y'. I will be trying to calibrate with a CH3I sample once I figure out what % v/v and in what solvent I'm supposed to use. - CaymanDave (Nov 02 '12 at 11:39)

Hi, sorry there was a typo in my previous post. If is 'f1coef' rather than 'f1coeff'. This variable is used to process the 2D depending upon the mode of acquisition (sensitivity enhanced/absolute value mode/etc..). This may or may not be the problem, but is worth a check. - Bharathwaj (Nov 09 '12 at 08:16)

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

powered by CNPROG