[time-nuts] 3-Cornered Hat fails with negative variances, is there anything I can to to improve?

[Erik Kaashoek]

After measuring and calculating the ADEV of three oscillators [1] the 
TAU files where created using Stable32 and the 3 TAU files where used to 
performed a 3-Cornered Hat variance analysis with this result:

3-Cornered Hat Separation of Variances
A-B Stability Data: RbvsRg.TAU
A-C Stability Data: OCXOvsRG.TAU
B-C Stability Data: OCXOvsRb.TAU
    Tau       Sigma A    Sigma B    Sigma C
2.000e-03  5.784e-10  Negative   4.732e-10
4.000e-03  2.957e-10  Negative   2.267e-10
8.000e-03  1.640e-10  Negative   1.123e-10
1.600e-02  1.096e-10  Negative   5.694e-11
3.200e-02Â  1.089e-10Â  3.416e-11Â  2.631e-11
6.400e-02Â  1.235e-10Â  3.859e-11Â  2.168e-11
1.280e-01Â  1.382e-10Â  3.659e-11Â  2.083e-11
2.560e-01Â  1.538e-10Â  5.818e-11Â  Negative
5.120e-01Â  1.718e-10Â  5.025e-11Â  Negative
1.024e+00Â  1.745e-10Â  5.221e-11Â  Negative
2.048e+00  1.822e-10  Negative   7.644e-11
4.096e+00  1.685e-10  Negative   7.813e-11
8.192e+00  1.519e-10  Negative   8.429e-11
1.638e+01  1.523e-10  Negative   8.295e-11

As far as I understand the negative variances are because the 
calculation method fails.
The method did calculate the variance for the least stable source (A)

Am I correct to assume the difference between B and C is too small for 
this method to separate their contribution?
Or is there anything I can/should to to separate B and C

[1] http://athome.kaashoek.com/time-nuts/DMTD/3-hat_input.png