Downamics, Andover (2026)

09/02/2026

Good data is about good preparation.

I usually take a couple of pretty well stocked trays of spare sensors and mounts, as well as a variety of nuts and bolts, strong double sided tapes and a huge amount of electrical tape and cable ties.

In the book, there is a chapter about being as prepared as possible regarding spares and consumables. Another chapter covers the testing protocols and structure of a day's testing that I've developed over time and the chapter on the equipment and sensors you need will depend on the goal of the testing. It's all well and good having a range of sensors and data channels, but if they're not installed properly or you don't have the correct methods to install them properly, then your data will be compromised. As with anything, getting the fundamentals right is the best first step you can make and the chapter on installation will mean you don't have to learn by making the same mistakes I did.

Data Driven Descents is getting closer and closer to being ready for a draft print which is really exciting. All being well, it will be released in March. Huge thanks to all the people who have placed preorders. It's blown my mind how many there have been. You can still preorder the book for the rest of February at the discounted price, if you haven't already.

04/02/2026

Steering dampers aren’t just down to “feel”… you can measure how effective they are. I have been running the .cc
CS.1 for around 15 months now and it's impressive.
Using frequency domain transfer analysis, we can easily see how effective the steering damper is at reducing vibrations from the front axle to the bars.

The result: ~17% reduction in vibration transfer from front axle to handlebars (0.5–500 Hz and up to ~23–27% reduction in key steering disturbance bands. This is exactly what you want from a steering damper. The transmissibility graph shows us how the steering damper has decoupled the front axle and the handlebars across the frequency range.

Less unwanted feedback = more stability, more grip, more confidence.

This isn't a sponsored post, I purchased my damper from Pademelon at full retail price, so this is a completely unbaised analysis.

31/01/2026

It's not a scientific test, and it's nothing new, but it is a proof of concept. Knowing an idea has been tried before shouldn't stop you from trying that idea yourself to understand it first hand.

Time to get some parts machined and actually do some proper testing.

If anyone knows of a good company to get parts machined, feel free to drop me a message.

30/01/2026

The Dorado rotor guard development is getting close to completion now. has been amazing at implementing changes and improvements to get this newest version ready for a test tomorrow.
Also another change to get some time on the .mtb brakes, and really starting to get to grips with them.
It's going to be a wet one at Forest of Dean tomorrow.

24/01/2026

Revisiting previous tests and analysing the data with new tools helps to either further prove the initial outcome or reveal other insights and/or outcomes that were previously not apparent.

In this case, using an expanded version of frequency domain transfer analysis to understand the effects of counter balancing wheels has been very interesting. The previous reports are then updated with revised analysis and outcomes based on the new findings.

New analysis tools applied to historic data can help to bring more value and deeper understanding to existing characteristics and the possible solutions to find further performance.

19/01/2026

I've put together a video on Patreon, which will be released on Wednesday, explaining how Frequency Domain Transfer Analysis can be used to understand how effective a lateral mass damper (LMD) is.

I have previously analysed it's contribution using accelerations in the time domain to show how, even on a gearbox bike, a lateral mass damper can reduce the amount of lateral movement at the BB/gearbox. This analysis also used FFT (frequency domain) to see how the amplitude of the frequencies is altered with a LMD. However, this new analysis shows the relationship between the rear axle (the excitation point) and the frame (the response point) and how the (LMD) can decouple the two, the energy distribution and the timing of the relationship.

I need to repeat this process on a non-gearbox bike(s) to see how the contribution differs, but also to understand how different the results may be on different bikes and with different rider inputs.

Link is in the bio.

12/01/2026

What does the Downamics IWM (Internal Wind Management) do? I used frequency domain transfer analysis to quantify its contribution (explained in last week's posts, but without gating). The .suspension does this for non-USD forks.

These slides are a comparison of No IWM vs IWM installed on a coil-sprung Dorado. There’s no air spring involved, the IWM is purely managing the air trapped in the upper tubes, which can influence how pressure builds and vents as the fork cycles.

For this comparison I’m looking at vertical acceleration at the front axle versus vertical acceleration at the handlebar, using frequency-domain tools.

Coherence (variable coupling) is the key metric for understanding how much of what the wheel is doing actually reaches the rider.

In the main suspension band (≈3–8 Hz), coherence is almost unchanged with the IWM installed. That tells us the IWM is not altering core suspension behaviour, which is exactly what we’d expect, it’s not a spring or a damper.

Where things change is at the higher frequency ranges. In the 60–200 Hz region, coherence is consistently lower with the IWM fitted. This means less high-frequency vibration is being transmitted from the axle to the bars, which is typically where buzz, harshness and hand fatigue live.

The PSD plots between the two runs are very similar in shape. This suggests the IWM isn’t dramatically changing how much energy exists in the system, the differences are subtle rather than transformational.

The Cross Phase plots largely overlap between runs. Where coherence is high (mainly at lower frequencies), the phase points cluster more tightly, indicating a more consistent and repeatable timing relationship. Where coherence is low, phase can look busy but isn’t something to over-interpret.

The IWM doesn’t produce a major improvement in suspension performance, which was never really the expectation.

It does reduce how higher-frequency vibration is coupled from the axle to the bars in this comparison.

That reduction won’t necessarily feel like “more support” or “more grip”, but it can translate into less harshness, less hand fatigue, & reduced cognitive load for the rider.

08/01/2026

How can we actually tell how effective a suspension system is, and what does "good" look like in terms of suspension performance?

These plots show coherence, cross-phase, and PSD analysis from a qualifying run at the 2024 Leogang World Cup. This is the same analysis process as the previous example, however this is showing suspension rather than using braking as a gate to analyse braking induced chassis dynamics.

By looking at how suspension displacement couples into frame acceleration across frequency, we can see:

- Strong, predictable coupling at low frequencies — where suspension control matters.

- Rapid decoupling beyond the working bandwidth — isolation doing its job.

- Clean phase behaviour only where coherence exists — no false conclusions.

- No evidence of high-frequency lock-in or instability.

This isn’t about “good” or “bad”, it’s about understanding how energy moves through the system, and where the suspension is controlling it versus isolating it. This is a trackside analysis method to aid in data driven decision making, but also as a way to understand the direction of the setup changes made based on proprietary software analysis during a day on track.

When setup decisions are based on this kind of insight, changes are made with intent, not assumption, but don't forget about the rider feedback!

06/01/2026

Just a bit of a follow-on post from the previous one regarding Brake-Gated Chassis Coupling Analysis, and another example of highlighting the affects of braking on the 🚲

Slide 1 – Lateral PSD (Rear Axle vs Frame)

The PSD shows us that the rear axle consistently carries more lateral energy than the frame across the full frequency range.
This is expected: the rear axle is closer to the primary source of lateral excitation — tyre contact, terrain inputs, and braking forces — while the frame acts as a downstream structure that filters and redistributes that energy.
Under braking, the PSD shape does not introduce new dominant frequencies.
Instead, braking changes how existing lateral energy is transmitted into the frame.

Slide 2 – Coherence (Rear Axle to Frame)

Where the system is dynamically coupled
Coherence highlights the frequencies where rear axle motion and frame motion are dynamically linked.
Three distinct regions stand out:
30–50 Hz: Strong, meaningful coupling
60–95 Hz: Secondary coupling band
145–200 Hz: High-frequency coupling that increases under braking
With brakes off, coupling is present but less “locked-in".
Under braking, coherence increases, indicating that braking stiffens the load path and transmits lateral dynamics more directly into the frame.
This is not added excitation, It's increased coupling.

Slide 3 – Cross Phase (Wrapped Phase)

Why coherence actually matters
The cross-phase plot confirms what coherence suggests.
In the same frequency bands (30–50 Hz and 60–95 Hz), we see:
Clear phase clustering
Consistent phase relationships
Tighter grouping under braking
Brake-on conditions show more phase consistency, especially in the 30–50 Hz band, meaning the rear axle and frame are moving in a more repeatable, synchronised way.
Brake-off conditions are visibly more scattered, less constrained, less deterministic.

Key takeaway

Braking does not create new lateral vibrations.
It changes the structural coupling of the system.
By increasing constraint through the rear triangle, braking allows existing lateral energy to propagate more directly into the frame, making certain frequency bands more dominant, more coherent, and more phase-stable.

06/01/2026

The Sego is ready for its close up. Time to get a .bike system installed and then some photos for the book to show good and bad examples of sensor installation, logger mounting and wiring.

Then it's time to get some laps on the new .mtb brakes and see if their performance is as good as their looks!

There's one more piece to this puzzle to come, and then it's a very full schedule of testing 👊

05/01/2026

The friction Reduction work that's been done on this DHX2 is even noticeable on the hand dyno, which is essential for running a shock through some tests after servicing work.

Apart from using a full blown dyno, this is the best way to check the compression and rebound damping functions as expected in a controlled, repeatable way.

It's not all graphs and squiggly lines.

03/01/2026

Another small way to help everything align as best as possible!
Treating the floating dropout and the axle to allow them to move as easily as possible helps with the lower legs and the axle alignment.

You can see in the first clip, when the lower leg is moved inward by a couple of millimetres, the stanchions drop. Giving the dropout and axle interface the best chance to find this position helps to remove this parasitic friction, and increase sensitivity.

When the axle and pinch bolts are torqued up in the correct order and with the fork compressed slightly, this small change can help the floating leg sit in the best position on the axle.

When this is done as part of our wider Friction Reduction treatment, the fork moves more efficiently and consistently — without needing to compromise spring or damping settings to mask friction that shouldn’t be there.

Downamics

09/02/2026

04/02/2026

31/01/2026

30/01/2026

24/01/2026

19/01/2026

12/01/2026

08/01/2026

06/01/2026

06/01/2026

05/01/2026

03/01/2026

Address

Website

Alerts

Contact The Practice

Shortcuts

Share