Frequency in very unstable

Hi I have asked about the Frequency In function of the T7 before for a hall effect speed sensor. My query today is related to a problem I am having with the stability of the frequency (1 / period measurement). Previously I had the sensor reading from a small DC fan with 2 metal contacts, this still works fine and I am able to read my frequency. When I connect my sensor up to its permanent position next to a trigger wheel which is on the brake for my dyno experiment I can't seem to get a stable signal. What I do get is a signal with ~10-20% error oscillations and frequent spikes of >100% of the signal frequency. Firstly I want to clarify my understanding of how the frequency in is read just to rule it out. Below are my configurations for the frequency in.

DIO_EF_CLOCK1_ENABLE            0

DIO_EF_CLOCK1_DIVISOR          32

DIO_EF_CLOCK1_ROLL_VALUE    0

DIO_EF_CLOCK1_ENABLE           1

DIO0_EF_ENABLE                       0

DIO0_EF_INDEX                         3

DIO0_EF_OPTIONS                     1

DIO0_EF_ENABLE                       1

I am reading from DIO0_EF_READ_A_F and getting the reciprocal of that for the frequency in Hz. I am using Clock1 as I require 2 different frequency PWM signals, the Clock2 frequency (14KHz) will give me a maximum period that is too small so I am left with Clock1. Previously I was using a 50Hz source on Clock1 but that means my minimum readable frequency would be 50Hz.

Clock Frequency = Core Frequency / Divisor

Max Period = Roll Value / Clock Frequency

Hence minimum frequency is just the reciprocal of the Max period. When I was running at 50Hz I noticed that when I increased the throttle of the engine the signal read from the speed sensor stabilised a little as soon as it was reading over 50Hz. This lead me to decrease the PWM frequency such that my new minimum readable frequency is 38Hz, which I thought might have been able to read my signal at is slowest. It has improved my readings but I still get very large spikes.

I intend to try lowering the frequency of my PWM further such that the minimum is a lot less than the original 50Hz but I still need my servo to operate at this frequency. 

Do you have any advice or ideas that could help at all? Would it be a bad idea to try lowering the frequency of my PWM any further?

Thanks,

Stuart.