Depends how they were connected - the benefit of an SPD will decline with longer connecting leads (i.e. away from the direct path from surge source to equipment to be protected) - usually 0.5m total is recommended, so if the total (of both L & N) from the SPD to the point where the other CU is connected (e.g. > 200mm to the Henley blocks) I'd say probably not adequately.

   - Andy.

Thanks Andy. Is that in the scenario where the SPD is in the CU further away than the CU without an SPD? If the SPD is in a CU nearest the origin and installed correctly less than 0.5m then this SPD will redirect the surge before it gets to the CU further away?

Agreed.

If you had a double-decker CU, a single SPD could be used. So if instead, the CUs are side-by-side, there is no reason in principle why a single SPD should not suffice, but the cables must still be short enough.

Jimi's second posting suggests that the second CU is supplied by a distribution circuit. My understanding is that an SPD at CU1 protects downstream circuits, but if the distribution circuit is too long (> 20 m?) another SPD may be required.

There will be some degree of spike reduction in almost any wiring arrangement with an SPD near when  compared to having no SPD, but its not a black and white thing, If you have the choice the suppression effect is best when the SPD iis nearer the source of the spikes you want suppressing.

As the current spikes have very fast rise times compared to 50Hz sinewave, the inductance of the wiring becomes far more important, hence the advice for short leads, and although the advice does not say so,  bringing the lines that carry the surge current through the SPD into close proximity so the magnetic field is partly cancelled, also helps, so cable or even a twisted bundle is better compared to a wide spaced loop of  singles.

Mike.