Lagunmatic 3516-SX Retrofit

andypugh wrote:

It might be worth creating a new 0V / Common for anything controlled by LinuxCNC.
Do you know what the operating current of that relay is? Is it within the current limits of the 7i77 outputs? (And it probably needs a protection diode, unless there is one in the relay)

You may be right about that and I'll have to look into how to do that. Good thought on the operating current - I couldn't find that exact spec but I did find the DC coil consumption which is 1.2 Watts = 50mA at 24v. I'm fairly sure this is the right spec but not 100% certain.

There is indeed a diode between the input wire [20] and common [1] at the relay. Thanks for watching out Andy.

Here are the photos I posted in larger format. Click for original size:

Original Wiring:

After changing transformer, spindle vfd, and connecting power:

How the 7i77 is currently wired:

Failed 7i77 wiring attempts:

giz

Ok, this sounds similar to a problem I had recently. I also found an old Lagunamatic VMX500 drawing I had and the 120VAC circuit was not tied to chassis ground on one side. Disconnect both the input power and output wiring leaving the chassis ground connected on both power supplies. Then using an ohm meter check the resistance between each input/output terminal and chassis ground. There should be no low ohms connections. On the input side you may get a high ohms reading because of the input filter. Also check between the input and output terminals, again there should be no low ohms connection. One of the power supplies may have the output 0V or return terminal connected to chassis ground, this is done to help with noise reduction. If this is the case then you will need to either get a new power supply with isolated outputs or find a way to break the 0V/chassis ground connection.

While you have the power supply connections removed it is a good time to check all the DC/AC control circuits for faults to chassis ground or open circuits. Using an ohm meter trace each control circuit and verify it is good. Find all the problems and fix them. If the 120VAC circuit is wired the way the drawing I have shows there should be no connection to chassis ground on either leg. Make sure all DC relays have working (correct polarity) flyback diodes across the coils to protect the Mesa card.

John

Thanks a ton for the lengthy explanation John - this will help a lot this weekend when I'll be at the machine next!

our machines do share neutral with 0v. Mine has been running like that for a year and a half now. Sorry i'm late to the party.

lunada wrote:

our machines do share neutral with 0v. Mine has been running like that for a year and a half now. Sorry i'm late to the party.

No worries! Any insights on how you grounded TB5 field power on the 7i77?

tb5? You mean TB2?
If you mean tb2, i came right off the 24v PS, and yes, it's still shared with neutral.

lunada wrote:

tb5? You mean TB2?
If you mean tb2, i came right off the 24v PS, and yes, it's still shared with neutral.

Yes, my mistake TB2. Thanks for confirming!

I ended up connecting a different power supply hoping it would make a difference - it did not. Same behavior as before - it will output 24v and power the 7i77 just fine until I connect the common wire to the V- side of the circuit. At that point, just like the original, the PS stops supplying DC voltage.

I brought in an extension cord and disconnected the input side of the 24v power supply from everything. I supplied power to the PS with 120v from the wall. Same behavior.

Schematics for the incoming power and outputs:

Giz

Looking at the drawings you posted I can see no reason why you can't make this work. All the power supplies you add need to be completely isolated input to output, this includes any chassis ground connections. Use your ohm meter to check them out. Did you have the safety ground connected in the extension cord you used for your test? If you disconnect all the outputs on your 7I77 does the power supply still fail when connected? A sketch of how you have wired your 7I77 to everything will help. The drawings do not show the servo power supply and wiring, this will also help.

John