You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to
Hi,
I have a Proform 770 EKG treadmill that at the moment with a problem
on the motor control board. The control board is called a model no.
MC-60, which was used in several different makes and models of
treadmill; as it turns out I'd scavenged a similar control board and
motor at the scrapyard a few years ago, and the treadmill seems to
have all functions restored when I swap this board in. However, I'd
like to repair the board, as I was hoping to use the spare one (and
motor) to
power a drill press or something shop related like that.
Hence, having a working circuit board to compare the non-working one
to, I'm hoping for some hints that will help isolate the problem
component. The motor is powered by 2 SCR's, which are not being
triggered. The trigger circuit has an opto-isolator MOC3052, which is
supposed to be activated by an op-amp on a nearby LM324, and I'm to
the point now that I think the op-amp is bad, from doing comparative
diode junction measurements on the two boards.
I wondering if I can confirm the faulty component's identity with my
DMM before I start replacing things - I don't have access to O-scopes
or anything fancier than a $30 DMM, so I wondering If getting
significantly lower readings on the suspect portion of the LM324,
would be a valid test (no short circuits, but 40 - or 50% lower mv's
on the diode-test scale.
So far I've only found a hand drawn schematic of the MC-60 board,
I kind of like the design of it as it has several LED's on it telling
what's working and what's not, I'll callcustomer support later in the
day to see if they can be of assistance.
Thanks in advance for any help,
Mike
hr(bob) hofmann@att.net
unread,
Jun 18, 2012, 9:07:46 AM6/18/12
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to
All I can say is good luck, treadmilll schematics are like hens's
teeth.
Jeff Liebermann
unread,
Jun 18, 2012, 10:47:14 AM6/18/12
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
>I wondering if I can confirm the faulty component's identity with my
>DMM before I start replacing things - I don't have access to O-scopes
>or anything fancier than a $30 DMM, so I wondering If getting
>significantly lower readings on the suspect portion of the LM324,
>would be a valid test (no short circuits, but 40 - or 50% lower mv's
>on the diode-test scale.
That might work if you had a schematic. The problem is that you can't
tell if you're measuring the LM324 or the components connected to it.
Looking at the photos, there are really only a few active components.
If you're desperate, try replacing the LM324 (with a socket).
>So far I've only found a hand drawn schematic of the MC-60 board,
>I kind of like the design of it as it has several LED's on it telling
>what's working and what's not, I'll callcustomer support later in the
>day to see if they can be of assistance.