This story begins back in March, when I got an amazingly crappy hetNOE spectrum from our 500 MHz spectrometer. The hetNOE is always a low-sensitivity spectrum, but for a 1 mM sample the signal I got was simply unacceptable. At first, I didn’t think much of this, for the simple reason that I thought the HX probe was in, and I’m used to getting poor signal-to-noise from that thing.
Except… except that everyone else was getting low-quality spectra too, no matter what probe was in. Finally one day Chris and Janice got almost no signal at all from an HSQC of a 1 mM monomeric protein, and that was the end. We had Sara test the magnet, and she found we had a four-fold reduction in signal to noise. To put things into perspective, we would have to take 16 times as many scans as before in order to achieve the same signal under these conditions. For a 1-hour HSQC, this is marginally acceptable. For a 2-day triple-resonance sequence, it is not. The new noise, strangely enough, came and went at odd times.
Al, because he knows all, immediately suspected a television signal, which he attributed to a rogue broadcaster somewhere out on Bear Hill. You see, an NMR spectrometer is really two things. It is a giant magnet, yes. But it is also a radiofrequency transmitter and receiver. The RF pulses induce a signal that the transmitter/receiver coil picks up. For our 500 MHz spectrometer, the primary signal we pick up is the proton signal at 499.75 MHz. This frequency is in a TV band.
Al was deflated, however, when Wlad tried to check the spectral band on his TV and found nothing. The search for the cause continued, with suspects ranging from the Brandeis student radio station to the construction crew radios to various parts of the magnet to secret government transmissions.
In the end, Al turned out to be right, and Wlad turned out to have a cheap TV. There was a channel broadcasting in this frequency range, but it was HDTV channel 18, a signal Wlad’s cheap TV cannot decode. Our spectrometer can’t decode it either, but we can sure see it. It knocks our signal down to an unrecoverable level. You can lose 75% of your signal for a small molecule and still be OK, but with proteins it’s a different story. The spectrometer is essentially useless to us in this condition.
And the worst part, the very worst part about this is that it’s a terrible channel! WMFP digital channel 18, broadcasting in the 494-500 MHz band, airs infomercials and “Gems TV”! Yes, that’s right, our scientific research has been derailed by a display of cubic zirconium jewelry in glorious high definition. Their antenna is located just south of us and we’re right in one of their strongest broadcast regions. Natural cures and cash-at-closing real estate ads are screwing us badly.
Hopefully, we’ll be able to boost the spectrometer’s frequency out of the channel 18 broadcast band to something like 500.13 MHz. However, there’s a Channel 19 (WGBH) broadcasting in the 500-506 MHz band as well; the spectrum analyzer shows a little gap between them that we can hopefully hit, and the WGBH signal is only half as strong as WMFP. Still, we may be screwed even if we boost the frequency. And there’s no way to shield the magnet, except maybe in the basement, but we’ll need to renovate the basement area in order to do that. So basically, we are screwed.