Inspired by Wildebus

wildebus said:
I have left my battery box (with battery, BMV, MPPT and Pi in it) alone since the 20th June...
Since then I have had Zero dropouts from the BMV communications to the Pi, but the MPPT comms are unreliable. The longest that has talked to the Pi is 9 hours without going AWOL. doing a Reset of the Pi via Remote Console once or twice will allow the MPPT to be seen again, but it will go away again a bit later

So one good, one bad.
Need to decide what to try next ....
Is it the cable? (swap VE.Direct plugs over - same device fails = cable)
Is it the Pi USB Ports? (swap USB plus over - same device fails = USB)
Could it actually be a fault with the MPPT if either of above happens? Or is the MPPT comms just more 'sensitive'?

Think I will move the two USBs over to the other unused USBs and see what happens ...
Click to expand...

How about using an Arduino to read the data from the MPPT. The Arduino should read the 3.x Volt Victron data as it's over 2.5 volts. Using SoftSerial to connect to the Victron. Then Serial.Print to the normal serial out which you can view in Serial Monitor. Add "time" to each write and then look to see when it drops.
 
xsilvergs said:
How about using an Arduino to read the data from the MPPT. The Arduino should read the 3.x Volt Victron data as it's over 2.5 volts. Using SoftSerial to connect to the Victron. Then Serial.Print to the normal serial out which you can view in Serial Monitor. Add "time" to each write and then look to see when it drops.
Click to expand...
Before I try a different OS or the like, I want to make sure I have reliable comms so I'm not ending up chasing my tail trying to troubleshoot multiple possibilities for an issue.
I can see the times the connection was lost by the sudden lost data in VRM, but there is no pattern or apparent logic to the occurances.
The top is BMV - data all the time from 20 June. The bottom is the MPPT - no lines means no comms. Only returns when I reboot.
1594036603792.png


Happens at any time (for example, just looked as typing this and it lost comms 13 minutes ago and will not be back until I remote reboot the Pi)
 
wildebus said:
Before I try a different OS or the like, I want to make sure I have reliable comms so I'm not ending up chasing my tail trying to troubleshoot multiple possibilities for an issue.
I can see the times the connection was lost by the sudden lost data in VRM, but there is no pattern or apparent logic to the occurances. Happens at any time (for example, just looked as typing this and it lose comms 13 minutes ago and will not be back until I remote reboot the Pi)
Click to expand...

My thought for using the Arduino was that you could see if the MPPT stopped sending data or if it never saw a drop the problem must be with the Pi.

As Victron spit data every second the Arduino could run a timer and if no data is received after 1.5 seconds it records it.
 
xsilvergs said:
My thought for using the Arduino was that you could see if the MPPT stopped sending data or if it never saw a drop the problem must be with the Pi.

As Victron spit data every second the Arduino could run a timer and if no data is received after 1.5 seconds it records it.
Click to expand...
that is a good point - is it the reciever or the transmitter at fault?

What I could do is put the 100/20 MPPT onto my regular system (Venus GX rather than Pi) and see how that goes (but was getting drops on the first set of cables I made).
we'll get there!
 
Instead of rebooting the Pi why don't you just unplug the USB lead and plug it back in? This is what I did for the first couple of days when I was having the same issue. Like I said once I switched the MPPT lead to a different USB port it's not lost the connection even after changing the 100/30 for the 100/50. Have you tried a genuine Victron lead? That would eliminate your homemade one. Looking on the internet indicates that a lot of people are using the Pi and Victron's own forums also show this is the case. It doesn't however indicate whether they are using homemade leads or the Victron ones.
What version of Venus OS are you running?

Regards,
Del
 
Look how often it disconnects. It is a lot easier to click a button then go outside to the workshop, open up the box unplug and plug in lead and come back each time.
Apart from which, the hands-off I think is a better test as it doesn't rely on fixing possible bad connections. It points to s/w I think (with Venus OS or MPPT)

Yesterday moved BOTH leads to different USB ports. BMV is still perfect. MPPT is still dropping.
Next step will be to swap VE.Direct plugs and see what happens.
 
wildebus said:
Look how often it disconnects. It is a lot easier to click a button then go outside to the workshop, open up the box unplug and plug in lead and come back each time.
Apart from which, the hands-off I think is a better test as it doesn't rely on fixing possible bad connections. It points to s/w I think (with Venus OS or MPPT)

Yesterday moved BOTH leads to different USB ports. BMV is still perfect. MPPT is still dropping.
Next step will be to swap VE.Direct plugs and see what happens.
Click to expand...
I see your point, it's just that it was easy for me as I'm in the van fulltime. I was thinking that rebooting maybe giving the ports a different address whereas unplugging would still use the same address :unsure: So are you still using homemade leads?

Regards,
Del
 
Good point about the port addresses maybe changing on reboots.
I have noticed - and am surprised - that the device ID does not change when the MPPT is found again after a reboot.
Still DIY leads. will shortly swap over the leads to the different devices. I think that could be a telling change....
 
Swapped the VE.Direct plugs over.....
The Device IDs also swapped over (I didn't reboot the Pi during this process, just moved the leads).

I guess that is due to the USB Ports each seeing now seeing a different device and allocating it a 'new' ID.
 
wildebus said:
that is a good point - is it the reciever or the transmitter at fault?

What I could do is put the 100/20 MPPT onto my regular system (Venus GX rather than Pi) and see how that goes (but was getting drops on the first set of cables I made).
we'll get there!
Click to expand...

I thought I posted this but obviously not and I can't multi task.

If you're intending running your own code on the Pi you the Arduino idea works the output could be level shifted back down to 3v3 and use a gpio pin
 
Re port numbering in Linux. To my knowledge when Linux boots it number the USB devices connected starting from 0 (zero), which device it starts numbering first is random, this has caused me issues in the past.
 
wildebus said:
Swapped the VE.Direct plugs over.....
The Device IDs also swapped over (I didn't reboot the Pi during this process, just moved the leads).

I guess that is due to the USB Ports each seeing now seeing a different device and allocating it a 'new' ID.
Click to expand...
When I unplugged the USB leads and plugged them back in (without a reboot) the VRM showed two MPPTs in the Device list, one it hadn't seen for ages and one it had seen 3 minutes ago, so I just deleted the one it hadn't seen for ages so I guess something changed when I did it? Or else I wouldn't have seen two MPPTs in the VRM :unsure:

Regards,
Del
 
This is what I currently have ....

1594139180168.png

I have decided to not delete the old IDs so I can keep a track better.

So 8 hours ago was when I switched the VE.Direct plugs over. Since then the MPPT has stayed connected but the BMV has lost connection twice - so by a process of elimination, it looks like I have a bad cable there
 
wildebus said:
This is what I currently have ....

View attachment 84123
I have decided to not delete the old IDs so I can keep a track better.

So 8 hours ago was when I switched the VE.Direct plugs over. Since then the MPPT has stayed connected but the BMV has lost connection twice - so by a process of elimination, it looks like I have a bad cable there
Click to expand...
Are you using USB extension leads? If so maybe swap them over and see if they are okay first?

Regards,
Del
 
nope. just a pair of modifed USB leads with the VE.Direct style connectors (JST 2.0) soldered on, so made as short as feasible.
I have 4 altogether so will swap the suspect one out in a couple of days (just going to monitor for a while first to check the same pattern happens)
 
A few more outages on the BMV, so got to be the cable.
Removed the cable few minutes ago and fitted one of the first ones I made. Will see how it goes!
 
wildebus said:
A few more outages on the BMV, so got to be the cable.
Removed the cable few minutes ago and fitted one of the first ones I made. Will see how it goes!
Click to expand...
So I fitted another cable to the BMV and that has been good. But the MPPT - which has been working great is now disappearing randomly despite not being touched.
The cost of the proper cables are not that much so I am throwing in the towel on these DIY cables as it is just not worth the time and hassle. We don't know what is in that lump on the end of the genuine Victron cables but there will be something there and whatever it is looks like it is important for reliable operation.

For messing around, use DIY cables, but for an installation that you want to rely on it communicating, I think the pukka cable has to be the way to go.
 
wildebus said:
So I fitted another cable to the BMV and that has been good. But the MPPT - which has been working great is now disappearing randomly despite not being touched.
The cost of the proper cables are not that much so I am throwing in the towel on these DIY cables as it is just not worth the time and hassle. We don't know what is in that lump on the end of the genuine Victron cables but there will be something there and whatever it is looks like it is important for reliable operation.

For messing around, use DIY cables, but for an installation that you want to rely on it communicating, I think the pukka cable has to be the way to go.
Click to expand...
I'll buy the Victron ones if and when my DIY ones pack in. I didn't make them to save money, I made them just to see if I could. Shame you've had so much trouble :(

Regards,
Del
Note added:
I've just checked and mine have been working without anything dropping out since I upgraded the MPPT controller on June 19th :cool:
 
alwaysared said:
I'll buy the Victron ones if and when my DIY ones pack in. I didn't make them to save money, I made them just to see if I could. Shame you've had so much trouble :(

Regards,
Del
Note added:
I've just checked and mine have been working without anything dropping out since I upgraded the MPPT controller on June 19th :cool:
Click to expand...
It is good yours is working and I am pleased I saw the post and tried it :) Just a shame it didn't work out as good in my case :(
Maybe I should post you mine to see if they work properly on your system? I think there is something borderline somewhere in the whole configuration. I could use one on my regular setup instead of a VE.Direct cable to see I guess?
 
You must log in or register to reply here.

Users who viewed this discussion (Total:0)

Back
Top