The board appears to be both sophisticated and flexible – for example, it can use armature voltage or a tachometer for speed feedback, it has several different input options to set the speed, and it has a control to increase motor agility up to the edge of instability, as well as the usual ‘IR comp’ control to optimise speed stability in the face of varying load.
It is a model 400 from Sprint Electric, and it was good to find out that the company that still thrives in Sussex UK, making products that include the great grand children of this controller.
Not only that, but a kind soul at Sprint responded to a begging email with circuit diagrams to show what loads the 0-10V speed control input.
A complicating factor for the motor speed controller project here is that all of model 400’s electronics float at mains line voltage (model 400i was the isolated version) – hence the need for an isolated circuit to bridge the gap between the available 5V pwm command signal and a analogue voltage riding on a potentially-lethal 240V sine wave.
Ebay and Amazon are awash with ‘isolated’ boards that convert pwm to 0-10V analogue, but they appear to be isolated in the sense that they an break an earth loop – great for machine control problem solving – but they are not isolated in the safety sense.
A note for the uninitiated coming across this blog: Safety-grade isolation is essential between live mains and any conductor that a human can touch. Lesser isolation might appear to do this job, but a century of fatal accidents suggests otherwise. Also: not all pwm to 0-10V converters are isolated at all. End note.
Reading around, some of these cheap pwm to 0-10V converter boards are non-linear, and not all have outputs that go all the way up to 10V.
The Sprint model 400 helpfully provides 240V-relative ‘0V’, ‘+10V’ (10mA max) and ‘+12V’ (10mA max) to aid the interface designer.
My control board – a BSMCE04U-PP from hymcu.com (aka ‘Bitsensor’ breakout) has an opto-isolated PWM output which uses an EL357N (rated as 3,750Vrms if it is a genuine Everlight) and has been designed with some thought, but clearance is as low as 2mm so it is only for breaking earth loops, and is not safety-rated.
EL357N is also not a fast opto-coupler, and the resistors around it have been picked to reduce power consumption rather than to increase speed – perhaps explaining user comments that it is non-linear unless driven with slow (1kHz did I hear?) PWM signals.
The hunt for safety-grade isolated pwm to 0-10V analogue continues….
Hats off to Sprint Electric.
