Baffled by intermittent running CH pump on Heatslave 15/19

[Fluke67]

Hi hoping someone can help with this as its beginning to drive me nuts. Boiler generally well serviced and working fine, no problems with DHW. Central heating recently began to occasionally fire up as normal and run until it reached temperature set by flow control thermostat at which point burner would turn off, do its little overun thing and then go completely silent. Realising that at this point CH water pump should be running and radiators should be hot not cold, I presumed pump not running, no hot flow to radiators, so burner off as water temp limit reached. Sharp tap to pump makes it whir into life, radiators get hot and burner fires up after a bit as ch water temp drops. This sometimes happens on initial demand for heat or randomly during a cycle. My boiler service guy has mentioned I might need a new pump soon, noisy, so I fit a new pump. Unfortunately still the same problem occurs. Tapping pump sometimes makes it run. Sludge, blockage stopping pump running is my initial thought but system relatively clean, dosed up and magnaclean type filter fitted. Put a plug on pump and plug it into a wall socket all works fine including boiler CH. Being mostly an electrician I get my meter out and discover only 75v ish on control board pins that suppy CH pump when its misbehaving and the requisite 240v when working as it should. I have a spare control board removed from a decommissioned but working heatslave so I swapped boards - same problem. I have also replaced the flow control thermostat like you do as one gets more baffled and could change the high limit thermostat, surely both control boards cant have the same fault?
Any ideas? Ive tried all the obvious, bypassing room thermostat, shorting out thermostats.
Thanks

 

[SimonG]

Does it have the built in clock? If so it will.be that.

 

[Fluke67]

Does it have the built in clock? If so it will.be that.

It doesn't have internal clock

Does it have the built in clock? If so it will.be that.

Does it have the built in clock? If so it will.be that.

It doesn't have the internal clock and i have bypassed all that. Too easy
I have actually been testing my spare board and have discovered open contacts, no continuity, on one relay that should be closed ( although to the eye it looks closed) like all the other 15 identical dpdt switches in all 4 relays, it even traces back to the pump terminals. If the board in use is the same I may have found the problem and although unlikely both boards do have the same fault. Intermittent partial contact in the relay could definitely be giving me 70ish volts at the pump instead of 240