AUC notes

Cell details

For maximum grip in the torque stand:
Clean inside of collet with ethanol to remove any Spinkote. Clean outside of cell
housings with ethanol, too.

The high speed velocity centerpieces from Spin Analytical can go up to 60K rpm but they hold ~70ul less volume than the regular ones. The timing on our fuge is close to the edge for these narrower sectors so data may have a "smile" to it.

One equilibrium cell housing, even with everything put together correctly, will have some overhang of one of the screw rings at one end. It still works.

Equilibrium runs can be performed using SV cells. The longer pathlength gives better resolution.

The pathlength for both types of centerpiece is 12 mm.

When using a needle to load a cell, be sure the needle tip doesn't touch a side wall. You get a wicking up effect which drives liquid into the filling tube, exactly what you're trying to avoid.

When using interference masks, the narrower slits go at the top of the housing.

Run details

The AUC's laser is at 675 nm.

When you change speeds during an equilibrium run, the computer will start a new folder for the new speed. Be sure to take 3K out of the method if you did a radial calibration first and uncheck that box. Otherwise, it'll do a radial cal. at all your speeds.

If you have meniscus-matching centerpieces and the menisci aren't matched at 3K rpm, run the fuge up to 10K briefly and that fixes it.

Spinning faster won't tighten peaks but it will increase the resolution between two nearby peaks.

You can change rotor temperature during a run but it will change at only 4°/hour.

The temperature at which radial calibration is done is irrelevant. Do it at 3K only. The instrument does compensate for rotor stretch.

An s value of 0.75 at 4° will be 1.15 at 20°.

To see all cells' scans at once, go to Window on the toolbar and choose Tile.

You don't need to have the instrument scan the counterbalance cell. Leave its boxes for Velocity, Absorbance, etc. unchecked.

If you're collecting both absorbance and interference data, stay at or below 40K rpm. There's a bug in the software which prevents the interference data from being saved if you're spinning faster. Also, do not ask to overlay the last 3 scans (unless you have 3 or fewer samples). A buffer fills fast and shuts down scanning.

If sample/buffer are in the wrong sectors, you can invert SE cells but not SV cells. Invert the data for the latter.

It takes ~1.2 min for the XL-I to do an absorbance scan at a single wavelength. The interference scans take 8 seconds. So if you want to add a delay between interference scans, take that scan time into account. The delay starts timing down after the scan of the first cell, even if you have seven cells present.

If you're measuring two different wavelengths in an absorbance run, have the software overlay the last 6 scans so you can monitor progress at both wavelengths simultaneously.

Fringe displacement lines do not have to overlay each other on the graph. They may be parallel for some of the run.

Pixel column shifting between runs is not unusual. The heat sink can shift with vacuum or speed.

Beckman recommends against stopping the fuge after the last scan. There's a bug there.

The laser delay will be different with air or liquid in the cells. For precision work, fill the cells with water and set up all the laser parameters at full speed. Remove water, put in samples, let equilibrate and spin with the water settings. You can even change cells after this as long as you don't change rotors.

For interference, once you've done the laser delay/radial cal, you can remove the CB cell and put a real sample in there. For absorbance, you need to leave the CB in there the whole time.

The temperature at which radial calibration is performed is irrelevant. Do it at 3K only, for absorbance runs. The fuge compensates for rotor stretch automatically.

The fuge always does Delay Calibration on cell 2 for the small rotor and cell 4 for the large rotor. Don't worry if it takes several minutes. If you get an error window, check that the software thinks it has the correct rotor. Check all 3 places. (listed on the wall)

If radial calibration is done at 3K before an SE run, start all over with a new method for your first speed. You can't take out "radial calibration before the first scan" on the fly, even though it looks that way on the screen.

If interference optics are used for a rotor, the radial cal. file for it can still be used for an absorbance run but not vice-versa.

It's okay to use an SV centerpiece across from the CB when doing an SE run. You can use either window type since the fuge only needs the 6.5 cm radial position.

Skipping the delay calibration will make for very noisy data. It is not recommended.

If repeating an interference run with the same samples resuspended, you don't need to re-do the laser delay and radial calibration. But you need to consider whether your molecules were affected by the first spin, perhaps dissociated.

You can use different optics on different cells if you've done your calibrations. For instance, do interference calibration on all samples, then you can ask for both absorbance and interference data from only one cell.

When spinning small peptides (~2K), use interference optics.

The range for Radial Step Size is 0.001 – 0.01 cm. The default is 0.003 cm.

Glycerol in samples should be avoided if possible. If not, keep it below 5% and don't spin at 4°. One user found that 2% was okay at 4° but 5% was not.

DTT in your sample could cause problems. See the RASMB messages from Borries Demeler and others from 6/6/06 on.

D2O and H2O weigh the same but D2O has much higher viscosity.

The laser in the AUC is 675 nm.

To loosen a red gasket from the edge of an aluminum centerpiece, sonicate it in water then pry at the gasket with the edge of plastic forceps or a Hamilton cleaning wire.

Recovering Samples
To recover SV samples, orient the housing with the screw ring facing you. Remove brass plug and red gaskets (carefully) from the R hand sector. Pipette up and down to resuspend your sample before you collect it.

To recover SE samples, loosen the screw ring with the part numbers upside down and to the left. Samples will be in the lower sectors, highest concentration to the left.

When looking at SE interference data in WinMatch, you'll see only one trace, not all three sectors (as you see with SE absorbance data).

Doing an extinction coefficient measurement

You can measure the extinction coefficient of a protein solution by doing a low-speed spin (3K rpm) using both absorbance and interference optics. (Do both types of radial calibration first.) Beckman has an application note for this: A-1815A. You need a synthetic boundary cell to do this.


If the vacuum goes up when you start a spin, that's oil outgassing and it will pass. If the vacuum goes to >100µ and stays there a while, you have a leak.

If things don't seem to be working at all correctly, check in three places that the setting for the number of rotor holes is correct. (See note taped to wall.) If this is incorrectly set, you'll get stuck in Delay Calibration or you won't see fringes.

If you see huge saw-tooth peaks when checking for whether equilibrium has been reached, you're looking at two different wavelengths and need to choose every other file.

If a window gets stuck in a window assembly, place an empty window holder under the stuck assembly and push directly on the window with something soft. (plastic forceps, finger)

A dark vertical line through the interference fringe display (and disturbance on the graphs) means the condensing lens cover is dirty.

If the temperature won't fall below 20.2°, you probably pushed the buttons in the incorrect order. Turn everything off, including the main power switch on the fuge and start again. (Fake spins begin with Start, not Vacuum.)

If it looks like one cell's data didn't get saved, it probably got defaulted to Notepad format so it won't have the .RA extension. Go to View, Folder Options, File Types and choose RA2. That should convert them all.

Wavy, variable interference fringes might mean that buffer leaked onto the windows. Lower contrast fringes will make for lower resolution (between two peaks). Downward fringes at the bottom of a sector might mean the cell is out of alignment in the rotor hole.

Negative absorbance and noisy data suggest that the cell is in 180° off.

If you have fringes from all cells except the CB, check the CB alignment.

A Z-shaped blip which doesn't move in an absorbance scan is the end of a hair which wasn't cut short enough.

Updated: 10/19/16