Jump to content

Supplementing DC rectifier with solar


Tacet

Recommended Posts

Hi all,

I'd like to pick a few brains, to see what the possibilities are.  The company I work for have a few DC rectifiers supplying -48 V to various battery banks and equipment.  We're unlikely to go for solar as we don't really have enough roof space.  However, I'm curious about the concept: how would you supplement an existing rectifier with solar?  The rectifiers are reasonably good at what they do, so you'd want the rectifier to retain control over the float voltage, temperature compensation, charge current limiting, LVD control, etc.

Simplest way: install hybrid AC/solar rectifier systems.  Various manufacturers make them, and they will do the job perfectly.  But replacing is expensive - supplementing should in theory be much cheaper.

Slightly less simple way: grid-tie inverters on the AC side.  However, that feels like you're simply adding unnecessary conversion losses and an additional layer of equipment.  The solution is readily available even in the retail market.

More efficient and cost effective way: supplement on the DC side at -48 V.  I've never seen solutions designed to do this, though.  I'm not even sure if it possible to do it without communication between the rectifier and the solar regulator, especially if it comes to charge current limiting. 

Any thoughts, solutions, ideas?...

I'm guessing that my preferred way is a pipedream.  If it was possible you'd probably see it punted widely as a way to easily increase solar capacity on the DC side, irrespective of what equipment is already installed.

Link to comment
Share on other sites

3 minutes ago, P1000 said:

In this scenario, what is the definition of a "rectifier" as you have used it here?

A device that takes as input either 3-phase or single phase AC, and rectifies it to -48 V DC.  On the DC side it will usually connect to both load and batteries, and the user will be able to set the float voltage as required by the batteries.  The rectifier can do temperature compensation and is able to limit the charge current to the batteries. 

Link to comment
Share on other sites

What is great about DC is that you can add different DC supplies together fairly easily.

So in your case I would simply get a number of PV panels and connect them to a 48V charge controller. This would then be connected to the batteries.

You would have to check the charging voltage but it should be a parallel system to the existing rectifier/charger.  

Link to comment
Share on other sites

14 hours ago, Richard Mackay said:

You would have to check the charging voltage but it should be a parallel system to the existing rectifier/charger.  

 

System sizes are scalable - I'm looking at systems from 2 kW to 30 kW.  I'm not worried about wire sizes - it is easy enough to bring your PV down to the charge controller at e.g. 250 V, and to install your charge controller next to the rectifier.

I don't think it is quite that easy.  If you simply parallel them and the output voltages are not exactly the same, the load will draw from the highest output voltage.  If that is rectifier, then the solar charge controller will idle.  If the charge controller tries to keep its output e.g. 0.01 V higher than that of the rectifier, the load will be supplied first by the charge controller and then by the rectifier.  We often parallel rectifier systems for redundancy purposes.  Even if the two rectifiers are from the same supplier, same model, in the same room (thus the same temperature) with the same configurations, you often see the load shift between the rectifiers under normal operating conditions.  Those minute differences between temperature readings and output voltages makes a difference. 

Add battery charging to the mix: you want the existing rectifier to control the battery charge voltage and to limit the charge current if necessary.  You still want to use your available solar energy to provide the load and charge the battery, but you don't want to overcharge the battery. 

The best idea of how to do this I have at the moment is to configure the charge controller independent of the rectifier, but with a float of e.g. 0.01 V higher than that of the rectifier. There may be times when the temperature compensation inaccuracies pushes the rectifier voltage higher than the charge controller voltage; it will be interesting to see how often that happens. 

Link to comment
Share on other sites

Why not just get a 3 phase grid-tied inverter and feed into the same 3 phase supply that is used for the rectifiers?

That way you can supplement the energy used by the rectifiers without needing to change anything on them or interfere with them in any way.

Link to comment
Share on other sites

1 hour ago, Tacet said:

 

System sizes are scalable - I'm looking at systems from 2 kW to 30 kW.  I'm not worried about wire sizes - it is easy enough to bring your PV down to the charge controller at e.g. 250 V, and to install your charge controller next to the rectifier.

I don't think it is quite that easy.  If you simply parallel them and the output voltages are not exactly the same, the load will draw from the highest output voltage.  If that is rectifier, then the solar charge controller will idle.  If the charge controller tries to keep its output e.g. 0.01 V higher than that of the rectifier, the load will be supplied first by the charge controller and then by the rectifier.  We often parallel rectifier systems for redundancy purposes.  Even if the two rectifiers are from the same supplier, same model, in the same room (thus the same temperature) with the same configurations, you often see the load shift between the rectifiers under normal operating conditions.  Those minute differences between temperature readings and output voltages makes a difference. 

Add battery charging to the mix: you want the existing rectifier to control the battery charge voltage and to limit the charge current if necessary.  You still want to use your available solar energy to provide the load and charge the battery, but you don't want to overcharge the battery. 

The best idea of how to do this I have at the moment is to configure the charge controller independent of the rectifier, but with a float of e.g. 0.01 V higher than that of the rectifier. There may be times when the temperature compensation inaccuracies pushes the rectifier voltage higher than the charge controller voltage; it will be interesting to see how often that happens. 

Good idea!

The fractionally higher voltage from the charge controller will allow the solar power to have priority over the existing rectifier.

With large charge currents you will also find the volt drop on the cables will probably bring both rectifiers into play.

I presume the 48V is for a short very high current application which is  costly if using a power supply??

Link to comment
Share on other sites

1 hour ago, Tacet said:

System sizes are scalable - I'm looking at systems from 2 kW to 30 kW.

30kW at 48V is jy gepik in die bol? Are you serious, 625Amperes? Really? What are you doing that requires this kind of current at that voltage? I canm't imagine the cable sizes required here. Some pictures please...

Yup hybrid inverter and feed your 'rectifiers' on the mains end via the inverter and when the sun don't shine then the Eksdom/Municipality can still provide, sounds like the best course of action.

Link to comment
Share on other sites

1 hour ago, Stanley said:

Why not just get a 3 phase grid-tied inverter and feed into the same 3 phase supply that is used for the rectifiers?

That's certainly the easiest solution, but not very efficient.  Instead of a single DC-DC conversion you now have DC-AC-DC.  Also, inverters tend to be more expensive than charge controllers, so it makes the ROI a more dubious deal. 

26 minutes ago, Kalahari Meerkat said:

Are you serious, 625Amperes? Really? What are you doing that requires this kind of current at that voltage? I canm't imagine the cable sizes required here. Some pictures please...

I try to keep work and forum separate, but this question piqued my curiosity, which is why I brought it here.  The industry is telecoms, so the main reason for operating at DC is to keep the equipment permanently on batteries.  A rectifier will typically be used to supply either multiple DBs or equipment racks, so in most cases you won't pull 625 A to a single load.  And yes, paralleled 150 mm2 or 185 mm2 cables does happen.  In some places busbars are used between the battery cabinets and the rectifiers. 

31 minutes ago, Richard Mackay said:

I presume the 48V is for a short very high current application which is  costly if using a power supply??

For high power solutions we prefer a UPS solution to carry the load until the site's EPS kicks in.  Unfortunately some of the equipment manufacturers are totally happy to have 18 kW devices that can only operate on 48 Vdc.

 

Link to comment
Share on other sites

Just now, Tacet said:

Also, inverters tend to be more expensive than charge controllers, so it makes the ROI a more dubious deal. 

I have found that pure grid-tied inverters (not hybrid inverters) are actually cheaper than charge controllers. Perhaps look at the Solis range of Inverters, they are very good and reasonably priced.

I agree that you will lose a little bit of efficiency, but not so much that I would be concerned.

Link to comment
Share on other sites

9 minutes ago, Tacet said:

The industry is telecoms

Ok, still seems a tad high, 2kW to maybe 5kW, I can imagine, more than that, yes, in a data centre... with lots of Intel processors turning electrons into heat 🙂 either way I am not aware of any direct solar charge controllers that would easily integrate into your infrastructure, its not exactly common and not every city block has 3 or 4 locations where this need exists, so, there's likely something out there that would fit, but I also suspect that the demand and volume for this product is so low, that going the inverter route may be a lot less expensive and not that much more inefficient.

Link to comment
Share on other sites

14 minutes ago, Stanley said:

I have found that pure grid-tied inverters (not hybrid inverters) are actually cheaper than charge controllers.

 

2 minutes ago, Kalahari Meerkat said:

Ok, still seems a tad high, 2kW to maybe 5kW, I can imagine, more than that, yes, in a data centre... with lots of Intel processors turning electrons into heat 🙂

...

...going the inverter route may be a lot less expensive and not that much more inefficient.

Some perspective - most operators are preparing for >15 kW loads for each mobile site when 5G comes.  I've seen small 19", 5U rectifiers that can do more than 20 kW.

I see pros and cons to mixing on the AC side or on the DC side.

Mixing on the AC side:

  • As you say, it may well be cheaper.
  • Extra conversion step - extra losses and extra things that can break.
  • A really nice pro - on a site with multiple small rectifiers, a single solar installation can benefit all the rectifiers as well as ancillary equipment such as aircons.
  • I don't think you'll be able to keep the inverters up if the mains fail, unless the solar installation/yield is larger than the site load.

Mixing on the DC side:

  • Simpler equipment, but a more difficult solution.
  • You don't have to worry about the AC side at all.  No backfeeding issues, no anti-islanding required.
  • In case of an AC fail, the solar will still contribute.

Thanks for the inputs, people!  You're making me rethink my stance on mixing on the AC side - there are more benefits than I realized. 

 

Link to comment
Share on other sites

16 minutes ago, Tacet said:

Mixing on the AC side:

If you have enough solar to provide the power needed, then a hybrid inverter with some small battery capacity, will keep on providing power, if the mains should fail, as long as the sun keeps on shining. The Sunsynk 5kW single phase hybrid I run here at home, self consumes around 70W/less than 100W and a smallish lead acid bank lets us use kW's of power, even if there's no mains about, during daylight hours.

Link to comment
Share on other sites

1 hour ago, Tacet said:

For high power solutions we prefer a UPS solution to carry the load until the site's EPS kicks in.  Unfortunately some of the equipment manufacturers are totally happy to have 18 kW devices that can only operate on 48 Vdc.

 

Are these batteries the 2V (1255Ah) lead acid cells?

Is there a changeover circuit for the 48V supply to be switched from the power supply (charger) to battery or is the load connected directly to the battery terminals?

Link to comment
Share on other sites

6 minutes ago, Kalahari Meerkat said:

If you have enough solar to provide the power needed, then a hybrid inverter with some small battery capacity, will keep on providing power, if the mains should fail, as long as the sun keeps on shining. The Sunsynk 5kW single phase hybrid I run here at home, self consumes around 70W/less than 100W and a smallish lead acid bank lets us use kW's of power, even if there's no mains about, during daylight hours.

Yep, that is the concept I use at home (my personal install is a Victron Multiplus II - works extremely well).  Most telco sites, though, has too little roof space to supply even close to the full load from solar.

1 minute ago, Richard Mackay said:

Are these batteries the 2V (1255Ah) lead acid cells?

Is there a changeover circuit for the 48V supply to be switched from the power supply (charger) to battery or is the load connected directly to the battery terminals?

The loads are usually connected to the batteries via an LVD that is controlled by the rectifier.  Under normal operation the LVD will be closed.

The batteries themselves are a fair range: 2 V VRLA, 12 V VRLA, lithium...

Link to comment
Share on other sites

17 minutes ago, Tacet said:

The batteries themselves are a fair range: 2 V VRLA, 12 V VRLA, lithium...

This makes implementing a PV solution on the DC side much more complicated. I would definitely go for the AC route. The only disadvantage being the loss of production during a power failure.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...