Connecting panels in parallel with Blocking Diodes

[victorroche]

I was told recently that panels in parallel should have Blocking Diodes in each branch. This is to prevent current flow from branches in high sunlight feeding current into those in shadow. Also we should consider Bypass Diodes in this.
I suppose the engineer is right, seeing that Solar Panels are active devices like batteries. For instance, if I connected a 12 volt battery across a 6 volt battery I would expect an almighty bang and high current flow, without a suitable load.
My main question is, why is this weakness never mentioned? Can these inter-panel blocking diodes be removed from each branch, thus saving me some extra costs. Of course, I do realize I require a blocking diode between these panels and the Charge Controller and Battery to prevent flow-back from the Battery.

I need some advice, please, guys!

[MarineLiner]

[Mike90250]

I was told recently that panels in parallel should have Blocking Diodes in each branch. This is to prevent current flow from branches in high sunlight feeding current into those in shadow. Also we should consider Bypass Diodes in this.
I suppose the engineer is right, seeing that Solar Panels are active devices like batteries. For instance, if I connected a 12 volt battery across a 6 volt battery I would expect an almighty bang and high current flow, without a suitable load.
My main question is, why is this weakness never mentioned? Can these inter-panel blocking diodes be removed from each branch, thus saving me some extra costs. Of course, I do realize I require a blocking diode between these panels and the Charge Controller and Battery to prevent flow-back from the Battery.

I need some advice, please, guys!

If you place your panels all in the sun, you don't need blocking diodes.
panels in the shade are useless.

Modern charge controllers all perform the blocking diode function, no need to add another one.

More than 2 strings in parallel, need FUSES on each string.

Bypass diodes are built into the panels.

Your engineer is in need of re-education, his advice is very out dated and somewhat wrong.

[victorroche]

Hi, Mike90250,

What you said make lots of sense. However, what happens if a temporary shadow is cast, or in the case of snow, etc. Will this cause a serious problem?
I suppose the moral of this story is to use proper Charge Controllers with relevant inbuilt diodes, etc.

[Mike90250]

no, fuses will protect if a failure occurs in a string, shaded parallel does not need diodes AFIK.

[hay4man]

Hi Mike,

I understand that you are very well versed in PV systems. I agree that if all the panels in the installation are guaranteed to not be shaded, then blocking diodes are unnecessary. I also agree that a shaded panel is useless. I just want to us to consider a few situations that might exist on a typical solar installation. Let's imagine a sailboat where two panels are installed but one panel is covered by a sail for several hours, or an installation on top of a house where multiple panels are installed, but one panel might be covered by snow or leaves (as Victor mentioned).

If one panel is shaded, and the other(s) has good sun, and the panels are connected in parallel without blocking diodes, I believe power will flow from the sun-covered panel directly to the shaded panel, and no (or little) power will reach the battery. This makes the system intolerant of shade. Also, the power that flows into the covered panel is being dissipated as heat in the covered panel. This will cause the covered panel to get hot, and possible shorten the lifetime of the panel. I'm interested to hear your thoughts on this.

Alexander

If you place your panels all in the sun, you don't need blocking diodes.
panels in the shade are useless.

Modern charge controllers all perform the blocking diode function, no need to add another one.

More than 2 strings in parallel, need FUSES on each string.

Bypass diodes are built into the panels.

Your engineer is in need of re-education, his advice is very out dated and somewhat wrong.

[inetdog]

Hi Mike,


If one panel is shaded, and the other(s) has good sun, and the panels are connected in parallel without blocking diodes, I believe power will flow from the sun-covered panel directly to the shaded panel, and no (or little) power will reach the battery. This makes the system intolerant of shade. Also, the power that flows into the covered panel is being dissipated as heat in the covered panel. This will cause the covered panel to get hot, and possible shorten the lifetime of the panel. I'm interested to hear your thoughts on this.

Alexander

One thing that you are not taking into consideration is that the Voc of a panel is not very dependent on the amount of light hitting it. The available current is. So unless the panels are not matched in construction, there will not be any voltage to drive current backwards into the shaded panel. Or at least the current will not be very large.
The common problem with shaded panels is that trying to force more current through them in the forward direction because of the voltage from unshaded panels in the same series string can damage the cells. That is handled by bypass diodes.

[Sunking]

If one panel is shaded, and the other(s) has good sun, and the panels are connected in parallel without blocking diodes, I believe power will flow from the sun-covered panel directly to the shaded panel, and no (or little) power will reach the battery. This makes the system intolerant of shade. Also, the power that flows into the covered panel is being dissipated as heat in the covered panel. This will cause the covered panel to get hot, and possible shorten the lifetime of the panel. I'm interested to hear your thoughts on this.

You have a valid point if we were talking about a DIY panel or some Grid Tied panels. But to my knowledge all manufactured battery panels have both bypass and blocking diodes built in the panel. GTI panels all have bypass diodes built in, but may or may not have blocking diodes built in. A simple Ohm meter test will verify the presence of a blocking diode.

[hay4man]

One thing that you are not taking into consideration is that the Voc of a panel is not very dependent on the amount of light hitting it. The available current is. So unless the panels are not matched in construction, there will not be any voltage to drive current backwards into the shaded panel. Or at least the current will not be very large.
The common problem with shaded panels is that trying to force more current through them in the forward direction because of the voltage from unshaded panels in the same series string can damage the cells. That is handled by bypass diodes.

I see what you're saying. I was operating under the assumption that a shaded panel Voc would be ~0V. I suppose the system wouldn't be affected very much if there is a lot of incident light or the panel is covered with something translucent.

Mod note - Links are generally not allowed

[inetdog]

I see what you're saying. I was operating under the assumption that a shaded panel Voc would be ~0V. I suppose the system wouldn't be affected very much if there is a lot of incident light or the panel is covered with something translucent.

One poster in another forum covered his panels with blankets and still measured a voltage which was about 3/4 of the Voc in full light. And even then the likely reason that the voltage measured was lower was that the voltmeter was pulling current.

Now if the panels are at wildly different temperatures, there might also be a problem, except for the fact that shaded panels are likely to be cooler and therefore have a higher Voc than the ones in sunlight.