For RX only
The length of each segment of the collinear antenna needs to be the same. The tolerance of differences decreases logarithmically with the increase in frequency. As example: HF 1/4 wave antenna segments (let's say 40 meter band [7 MHz]) is 10.5 meters. A 1 cm more or less is not going to make a dramatic impact on the resonant frequency. But if the frequency is 7 GHz, then a 1/4 wave length will be 10.7 cm, here a deviation of 1 cm will be disastrous. Moral to the story: the higher the frequency = SIZE MATTERS accuracy is very important and becomes crucial.
If you cut the segments to be exactly a 1/4 wave length (as they should be in theory) it infers that when combining the elements during the construction of the collinear antenna, the distance between the elements (theoretically) should be 0 mm. In practice this not possible and there should be insulation to prevent unwanted short circuits, specially with braided coax cable that is frayed. I use 0.18mm insulation tape. I was informed by an eager enthusiast that the antenna did not work at all on the selected frequency, but did many MHz lower in the band. Upon further questioning, nylon (clothing) buttons about 4mm thick where used as insulation between the elements for an ADSB antenna (1GHz+). Brilliant idea and solution! But then do compensate for extra insulation length when cutting the elements to size. Rule of thumb: 1/2 the thickness of the button removed (cut or measured shorter) on each side of an element of the antenna that has contact with a 'button' (of 6mm = 3mm) should do the trick and keep the selected frequency correct for the antenna construction.
For antennas that are used only for receiving... The closer you get to resonance the better, however, being a bit off frequency is not a total disaster. The antenna will still receive very well as it will act as a band pass filter, provided there are no strong TX stations on the actual resonant frequency of the antenna.
For RX and TX
If you plan to use the antenna in a PVC pipe and outside in an environment where temperatures fluctuate that could cause the due point inside the PVC pipe to be reached..then. (1). Tune the antenna IN THE PVC PIPE, and preferably in winter with low humidity before you seal the pipe. The enclosure does affect the resonance, as close by objects that are not (preferably a vacuum) air (the PVC pipe thus), will cause electro magnetic deviations in the near field environment of the antenna.
(2) Putting RF power into the antenna will cause a part of the coax feed line to become part of the antenna, hence the ferrite at approximately 1/4 wave length (ultimate tuning matching stub). We don't want RF on the feed line. This ferrite is also your fine tunning device to get you onto the seat spot of the antenna.
(3) If you fear dew point and water condensation in the pipe, pop in a small satchel of silicon beads at the base (where the feed line connects), it will help keep the antenna on frequency in varying weather conditions. There are other solutions with valves, but this has a price attached.
This antenna is very interesting because of the potential amplification (gain) of the received radio signals for a specific frequency, however it is a real challenge to get right the first time!! It's all about the materials, environment and accuracy of construction, and even then mother nature has a way of making things go sideways sometimes. Like a rusty screw on the antenna mounting. Believe me it can really be an issue!
Failing the ferrite on the feed line as matching stub (I don't have ferrite!), a small 'earth plane', or artificial ground (see 1/4 wave antenna design) can also work, however the 45 degrees of a 1/4 wave does not apply, you will need to experiment.Also the length will be less then 1/4 wave.