Relay dropping out momentarily

**newkirk** · 04-27-2010, 07:01 PM

Originally Posted by newkirk

Later tonight I may be in-dash in the car, if so I'll try to set it up powered by the car battery and see what happens.

Well I don't have a final verdict yet, since I was in-dash to remove the temporarily-mounted PC, but I had the FB running off car battery for several minutes, during which time the relay didn't click a single time. (it still buzzes, fairly loud, but I'm hopeful that once embedded deep in the dashboard I won't be able to hear that)

j

**newkirk** · 04-27-2010, 08:45 PM

Further info:

When plugged into the same original power source, the relay no longer drops out!! Perhaps the fix is to blow up the voltage regulator?

More seriously, could someone who is experiencing this issue try something: Set your FB to power via USB, so that the (still-connected) 12V is used only on the outputs instead of powering the brain itself, and see if you still experience dropouts or not.

j

**acbusdriver** · 05-09-2010, 04:32 PM

I got my replacement brain... does the same thing. I've tried multiple 12V power sources, no difference.

As newkirk requested somebody try, I did try switching the power jumpers over to USB, while leaving the 12V plugged in. It operates exactly the same. just that split second drop out that gives the annoying click.

maybe i should blow up the voltage regulator and see if that helps

**newkirk** · 05-09-2010, 06:32 PM

Originally Posted by acbusdriver

I got my replacement brain... does the same thing. I've tried multiple 12V power sources, no difference.

As newkirk requested somebody try, I did try switching the power jumpers over to USB, while leaving the 12V plugged in. It operates exactly the same. just that split second drop out that gives the annoying click.

maybe i should blow up the voltage regulator and see if that helps

Depressing news. I have to amend my previous statement anyway: It DOES still drop out momentarily. But it doesn't do it with the same frequency on each output. Specifically, outputs 1-4 and 14-16 do it most often, sometimes it will click 4-5 times in rapid succession every ten seconds or so. Meanwhile, connectors toward the middle of the board (like 8-10) only do it once every few minutes or longer. Still not acceptable for most uses.

When dropping out, the indicator LEDs flicker as well, so I've set all outputs ON and watched for several minutes - There's a clear pattern where the extreme outputs (1-4 and 12-16) flicker most often, and usually together. (well, per end - when 1 goes 2-4 usually accompany it, when 16 goes, 13-15 usually do) When this happens, output 1 for example might click (and flicker) five times, same for output 2, but 3 and 4 might only do it three times. I've attached a relay to each output for a time and observed that it drops out in sync with the flicker, as expected.

BTW, a disclaimer: mine is a Defective Brain. The only defect I've explicitly identified is that the LED for output 9 doesn't light. In other circumstances I would probably consider the momentary dropouts part of the 'defective' classification, but I presume others experiencing this are using the 'real' V4 fusionbrains, and in addition the product description indicates that at least some outputs and inputs will work regardless of defective status.

Can't we surmount this with a capacitor on the output?

j

**acbusdriver** · 05-09-2010, 07:53 PM

Originally Posted by newkirk

Can't we surmount this with a capacitor on the output?

j

This option did cross my mind. I might give it a try. I hope the problem can be solved sometime otherwise, but that will probably work in the meantime. I would like to use a pwm output as well, I hope that the dropout problem won't affect that.

As you said, I have also noticed that the problem is based on the outputs directly, not the relays, it's just the relays that make the problem noticeable.

**2k1Toaster** · 05-09-2010, 10:53 PM

How many relays are you powering at once?

You can try shorting your thermal fuse which will allow any number of relays at a time. Currently it is setup to allow only a minimal amount without modification so that you dont blow them up

**acbusdriver** · 05-09-2010, 11:06 PM

I've only been powering 1 relay to test it. I've tried powering multiple relays at once (usually 3 or 4, I tried 6 once and was treated to a full chorus of clicks) and the problem doesn't seem to change. I haven't gone to the extent of checking them as newkirk did with seeing if some of the outputs are worse than others but the fact is that they all do it. I'll try shorting the thermal fuse.

**ACCLR8N** · 05-10-2010, 07:29 PM

I was having issues with commands from RR with outputs dropping out in the software (virtual brain). I thought it had something to do with the timers. I corrected it by adding a 3 or 4 logic count for each on vote command.

I haven't noticed relay clicking, but I only have a few 1A micro relays hooked up.

**NickSVR4** · 05-13-2010, 01:59 PM

Originally Posted by newkirk

Depressing news. I have to amend my previous statement anyway: It DOES still drop out momentarily. But it doesn't do it with the same frequency on each output. Specifically, outputs 1-4 and 14-16 do it most often, sometimes it will click 4-5 times in rapid succession every ten seconds or so. Meanwhile, connectors toward the middle of the board (like 8-10) only do it once every few minutes or longer. Still not acceptable for most uses.

When dropping out, the indicator LEDs flicker as well, so I've set all outputs ON and watched for several minutes - There's a clear pattern where the extreme outputs (1-4 and 12-16) flicker most often, and usually together. (well, per end - when 1 goes 2-4 usually accompany it, when 16 goes, 13-15 usually do) When this happens, output 1 for example might click (and flicker) five times, same for output 2, but 3 and 4 might only do it three times. I've attached a relay to each output for a time and observed that it drops out in sync with the flicker, as expected.

BTW, a disclaimer: mine is a Defective Brain. The only defect I've explicitly identified is that the LED for output 9 doesn't light. In other circumstances I would probably consider the momentary dropouts part of the 'defective' classification, but I presume others experiencing this are using the 'real' V4 fusionbrains, and in addition the product description indicates that at least some outputs and inputs will work regardless of defective status.

Can't we surmount this with a capacitor on the output?

j

Without relays hooked up, I have noticed this drop off with the LEDs at well. Is this something to be concerned about right now? With all output LEDs on, then generally all will very briefly turn off then back on (Flicker). This varies in time, sometimes being every couple seconds, to a couple times per second.

Is this due to a timer issue in MDX or is this a hardware issue?

**newkirk** · 05-13-2010, 04:19 PM

Originally Posted by NickSVR4

Without relays hooked up, I have noticed this drop off with the LEDs at well. Is this something to be concerned about right now? With all output LEDs on, then generally all will very briefly turn off then back on (Flicker). This varies in time, sometimes being every couple seconds, to a couple times per second.

Is this due to a timer issue in MDX or is this a hardware issue?

MDX shouldn't have anything to do with it - for normal toggle-on toggle-off operation, it sends a message to the FB telling it what to do, and that's it - no constant intervention between MDX and the FB. PWM and even a selection of timed-duration settings are all handled within the FB itself, MDX only needs to send messages to it telling it when a setting needs to change.

My gut feeling is that you're looking at the same issue we're discussing, but don't kill me if I'm wrong...

j