Anybody using generator linked in via rectifier?

[macafrican]

For tying in a generator to a system, kicking in when battery SOC gets a bit low, has anybody gone DC tied with a rectifier instead of normal generator AC route?

In my case I want to keep grid connection if it is around, but severely capped in kVA.    Synchronizing generator and grid is tricky and some inverters don’t like it plus many installers just say no quoting nonsensical rules.

So alternative method is gennie : rectifier : DC Combiner.    That DC Combiner is the junction where inverter picks up DC connection and to which the battery banks are also connected.    Concept is

-Battery gets to target example 30%

- Gennie fires up and supplies DC to the Combiner

- depending solar and grid limit and loads the inverter either charges battery or draws it down much slower.

-Battery gets to upper target SOC say 75% and shuts down.

The nice part is it is fairly clean install, the gennie need not be sized to peak inverter size and there is absolutely no change-over on the AC bus side.

[WannabeSolarSparky]

You would ideally use something like this:
https://eg4electronics.com/wp-content/uploads/2024/05/Chargeverter-5kW-Spec-Sheet.pdf

But not really seen anything similar for sale in SA yet.

Else with a bit of electronics knowledge you could simply make a converter/rectifier to output required DC voltage and connect direct to battery bus.
Which is pretty much what the chargeverter does.

[BritishRacingGreen]

1 hour ago, macafrican said:

For tying in a generator to a system, kicking in when battery SOC gets a bit low, has anybody gone DC tied with a rectifier instead of normal generator AC route?

In my case I want to keep grid connection if it is around, but severely capped in kVA.    Synchronizing generator and grid is tricky and some inverters don’t like it plus many installers just say no quoting nonsensical rules.

So alternative method is gennie : rectifier : DC Combiner.    That DC Combiner is the junction where inverter picks up DC connection and to which the battery banks are also connected.    Concept is

-Battery gets to target example 30%

- Gennie fires up and supplies DC to the Combiner

- depending solar and grid limit and loads the inverter either charges battery or draws it down much slower.

-Battery gets to upper target SOC say 75% and shuts down.

The nice part is it is fairly clean install, the gennie need not be sized to peak inverter size and there is absolutely no change-over on the AC bus side.

DC coupling solution for the high voltage batteries might be tricky. I know Atess has a rectifier cabinet for DC coupling, but thats probably very expensive. 

If its LV (48V) I would  consider a  hybrid inverter for this purpose. 

[BritishRacingGreen]

7 minutes ago, BritishRacingGreen said:

DC coupling solution for the high voltage batteries might be tricky. I know Atess has a rectifier cabinet for DC coupling, but thats probably very expensive. 

If its LV (48V) I would  consider a  hybrid inverter for this purpose. 

Slightly offtopic, with a rectifier dc couplled solution where you connect your mains also to such rectifier, you get  0mS transfer to load when mains fail, very solid and reliable solution for critical loads. Only downside is a bit of loss of efficiency. 

[macafrican]

There are actually some people making rectifiers in SA - there are all sorts of industrial applications where people need large DC power supply.   Those normally assume grid AC : rectifier = DC output.

For mine I need high voltage (775V) and yes the Atess is $$$ but I figure in a home setup or small office setup this would ALSO be far preferable than doing the ATS/Bypass usual trick.   Was wondering what people’s experience is with generator:rectifier and whether have to be careful about generator type, alternator, voltage regulator, governor, etc.

As I have it, the rectifier is set up to present to the generator as a constant load being taken.

My other concern is that DC Combiner added to the mix.    Basically large DC bus that has the input from the rectifier but also has cables to battery and cables to the battery inverter.     Injecting DC there is presumably just same as DC that would have come from grid/solar via the battery inverter / PCS, so the BMS presents either charge demand or takes recharge from that DC busbar?

I would set mine up that BMS max charge rate is a little below the rectifier max output.    Else the low SOC battery might ask for more kW than the gennie : rectifier can supply and that might get messy.

[macafrican]

3 hours ago, WannabeSolarSparky said:

You would ideally use something like this:
https://eg4electronics.com/wp-content/uploads/2024/05/Chargeverter-5kW-Spec-Sheet.pdf

But not really seen anything similar for sale in SA yet.

Else with a bit of electronics knowledge you could simply make a converter/rectifier to output required DC voltage and connect direct to battery bus.
Which is pretty much what the chargeverter does.

Thanks - I want to keep charging / discharging control under the BMS and PCS.    My worry with a battery charger is it then messing with the BMS.   So just pure DC supply whatever kVA until get signal to stop gennie.

[BritishRacingGreen]

2 hours ago, macafrican said:

Was wondering what people’s experience is with generator:rectifier and whether have to be careful about generator type, alternator, voltage regulator, governor, etc.

As I have it, the rectifier is set up to present to the generator as a constant load being taken

I dont have field experience here, but I can assure you one thing that the battery is now a perfect filter component as far as the generator is concerned. It does not get better than this. Your lithium battery has got such a low source impedance that it will just suck up any surge that the load wishes to impose within design parameters of the system.  The generator is therefore abstracted from abrupt load variations. 

 

2 hours ago, macafrican said:

My other concern is that DC Combiner added to the mix.    Basically large DC bus that has the input from the rectifier but also has cables to battery and cables to the battery inverter.     Injecting DC there is presumably just same as DC that would have come from grid/solar via the battery inverter / PCS, so the BMS presents either charge demand or takes recharge from that DC busbar?

I would set mine up that BMS max charge rate is a little below the rectifier max output.    Else the low SOC battery might ask for more kW than the gennie : rectifier can supply and that might get messy.

DC blending of sources are just the perfect way nature intended. 

I would actually suggest the opposite in that you set the rectifier to a voltage that is  lower than the inverter charger. This is maybe required as to not interfere with the transaction between bms and inverter. A pedantic bms or pedantic inverter might abort operation if the actual charge voltage is higher than that negotiated. But this I have not witnessed in practice. Maybe @Coulomb can give us his thoughts here. 

 

 

Edited September 26 by BritishRacingGreen

[BritishRacingGreen]

21 minutes ago, BritishRacingGreen said:

I dont have field experience here, but I can assure you one thing that the battery is now a perfect filter component as far as the generator is concerned. It does not get better than this. Your lithium battery has got such a low source impedance that it will just suck up any surge that the load wishes to impose within design parameters of the system.  The generator is therefore abstracted from abrupt load variations. 

 

DC blending of sources are just the perfect way nature intended. 

I would actually suggest the opposite in that you set the rectifier to a voltage that is  lower than the inverter charger. This is maybe required as to not interfere with the transaction between bms and inverter. A pedantic bms or pedantic inverter might abort operation if the actual charge voltage is higher than that negotiated. But this I have not witnessed in practice. Maybe @Coulomb can give us his thoughts here. 

 

 

One drawback  is that generator must have enough capacity to both charge a depleted battery as well as supplying the load demand. If not the rectifier must have suitable current limiting capability. 

In an ac coupled arrangement the inverter can be configured to impose a lower charge current limit when it is in generator mode. 

Edited September 26 by BritishRacingGreen

[Scorp007]

1 hour ago, BritishRacingGreen said:

One drawback  is that generator must have enough capacity to both charge a depleted battery as well as supplying the load demand. If not the rectifier must have suitable current limiting capability. 

In an ac coupled arrangement the inverter can be configured to impose a lower charge current limit when it is in generator mode. 

 

1 hour ago, BritishRacingGreen said:

DC blending of sources are just the perfect way nature intended. 

 Maybe @Coulomb can give us his thoughts here. 

 

 

From a previous reply from @Coulomb he did indicate that the extra source to charge at a lower level to allow the inverter to do the last bit in topping up. 

Edited September 26 by Scorp007

[WannabeSolarSparky]

3 hours ago, macafrican said:

Thanks - I want to keep charging / discharging control under the BMS and PCS

You set the charger voltages below the bms voltages that way the BMS stays in charge/control of the charging and the generator will only provide what the charger requests.

[Tinbum]

I was going to have a look at this but with lots of other things to do haven't had time. I bought a couple of 48v Hybrid starter/alternators but, as of yet, haven't found the pin configuration for its control and reading the CAN protocol. I'd want the bms being in total control of it.

Edited September 26 by Tinbum

[Steve87]

@macafrican Atess have a RTF 600 unit that works with the PCS. It's a 600kW unit & because it's DC coupled there is no switching time because the load is always online. The PCS & the RTF both share the battery. There are advantages to this system like no switching time but the disadvantage is that you cannot export power because the PCS is always in off grid mode. 

If your system made use of the bypass cabinet & now the 20ms switch time is a problem & you were using a Grid tie system & AC Coupling you can still AC couple on the load side, however, you will need an EMS gateway that entertains both the PCS & Grid tie because when the batteries are full the Grid tie inverters need to throttle down. The PCS does do frequency shifting, however, the question remains whether that only works on the Bypass cabinet AC Coupling architecture??

Atess themselves have only comissioned one in the country so it's a new kid on the block. However from my technical information you choose one or the other between DC coupling RTF & the Bypass system. The AC coupling is still however possible with an EMS a real possibility or using their PBD250/350 if that is what was chosen from the beginning. 

I will hopefully commission my first one using the RTF in the coming months...

[Jacques Ester]

You can contact Semikron.

They specialize in rectifiers.

They build Rail and trolley systems for the industry.

They should be able to assist in any rectifier that you need.

They used to build 3 phase 1600V 2000A rectifiers for our mine.

 

[macafrican]

On 2024/09/26 at 9:14 PM, Steve87 said:

@macafrican Atess have a RTF 600 unit that works with the PCS. It's a 600kW unit & because it's DC coupled there is no switching time because the load is always online. The PCS & the RTF both share the battery. There are advantages to this system like no switching time but the disadvantage is that you cannot export power because the PCS is always in off grid mode. 

If your system made use of the bypass cabinet & now the 20ms switch time is a problem & you were using a Grid tie system & AC Coupling you can still AC couple on the load side, however, you will need an EMS gateway that entertains both the PCS & Grid tie because when the batteries are full the Grid tie inverters need to throttle down. The PCS does do frequency shifting, however, the question remains whether that only works on the Bypass cabinet AC Coupling architecture??

Atess themselves have only comissioned one in the country so it's a new kid on the block. However from my technical information you choose one or the other between DC coupling RTF & the Bypass system. The AC coupling is still however possible with an EMS a real possibility or using their PBD250/350 if that is what was chosen from the beginning. 

I will hopefully commission my first one using the RTF in the coming months...

Atess have a 300 rectifier as well.

I have Atess PCS, isolating transformer and bypass cabinet plus IES battery and a lot of solar, then grid and generator.   So an AC coupled smartgrid.    Most all is perfect.   My use case is grid throttling because I get screwed R440/kVA for peak half hour kVA in month.   I throttle grid at 90, loads reach around 300 but solar and battery takes care of that.    All of this now works fine, though had to EMS and PPC to make solar and exports work safely as Atess can basically only work with Atess solar gear.

I need solution for when have bad solar, maybe some loadshed, battery getting low and need gennie to help for a while.

My problem is Atess’s idea of generator integration is a COMPLETE JOKE.    (1) despite what the manual says it cannot run grid and gennie concurrently - apparently a translation error in the manual where it says those three lights will be on when grid and gennie supplying. (2) when gennie running, Atess switches to fixed kW battery charge rate - really, no jokes (2) the contactors in bypass cabinet to change over between grid and gennie are so bad that it blows PCB of aircons and chillers (3) standard way Atess cannot run solar during generator.    They suggested I ugrade my 420 Scania to 800

They are afraid of the obvious method : run sync mode gennie output to my load busbars.   Harolds guy told me it is anyway not possible to sync gennie and AC (even though hundreds of thousands of people do so every day).

That leaves me with take gennie output away from bypass cabinet, send that to rectifier that then supplies DC to the DC combiner.     So I can run gennie 300, grid 90, whatever solar I have, handle loads and then the rest charges battery subject to my own 300 max charge rate, get to 75% SOC when gennie stops.

So in essence a hybrid of DC and AC coupled smartgrid.

[macafrican]

On 2024/09/27 at 7:34 AM, Jacques Ester said:

You can contact Semikron.

They specialize in rectifiers.

They build Rail and trolley systems for the industry.

They should be able to assist in any rectifier that you need.

They used to build 3 phase 1600V 2000A rectifiers for our mine.

 

Great!   I’ll try them for sure.

[macafrican]

On 2024/09/26 at 2:47 PM, BritishRacingGreen said:

One drawback  is that generator must have enough capacity to both charge a depleted battery as well as supplying the load demand. If not the rectifier must have suitable current limiting capability. 

In an ac coupled arrangement the inverter can be configured to impose a lower charge current limit when it is in generator mode. 

I get what you mean, but in reality not needed.    Simplistically, view it same like you have a 15kW system but only 9kW of solar.    My rectifier capacity needs to be close enough to load to be able to at least slow down my rate of battery discharge (if grid down and solar is almost nothing).    In reality my loads are in any event mainly daytime and my solar is 150% of loads so even cloudy day I have SOME solar.     Ideally, I have my grid capped input, solar input, gennie input, load output and the balance is either charging or discharging.

[macafrican]

On 2024/09/26 at 5:44 PM, Tinbum said:

I was going to have a look at this but with lots of other things to do haven't had time. I bought a couple of 48v Hybrid starter/alternators but, as of yet, haven't found the pin configuration for its control and reading the CAN protocol. I'd want the bms being in total control of it.

Agree on keeping the BMS in control!

Most small generators in homes or SME are themselves dubious quality and/or difficult to integrate seamlessly.    I like how the rectifier route is seamless on the AC in terms of no possible interruption, and the rectifier can sit as constant load on generator hopefully at its efficient output level until battery gets to target and then stop.

[macafrican]

On 2024/09/26 at 3:50 PM, Scorp007 said:

 

From a previous reply from @Coulomb he did indicate that the extra source to charge at a lower level to allow the inverter to do the last bit in topping up. 

I don’t get that part about rectifier lower output voltage?      Eg in my case the DC connection is 750V but the really critical part is that each cluster BMS stops a module in that cluster if a cel in a module exceeds I think it is 3.55V.    In a home setup the supply to BMS would be about 52V (?) and the BMS then handles voltage per cel and balancing.

I’m a bean counter not electrician, so on thin ice when it gets to BMS

[Steve87]

1 hour ago, macafrican said:

standard way Atess cannot run solar during generator.  

Hi @macafrican having commissioned more than 60 ATESS Plants in many different configurations, some with Genies, some with 2 Genies: This statement of the ATESS is incorrect. You can have Gene Power or Grid power never both Synchronized & powering the same AC bus. There is no system with reasonable Hybrid technology that is doing that on the market. That's like like a Sunsynk with Grid power in peak shaving mode & then running a Generator at the same time & they are able to blend power together. It's just not possible because the two sources are out of phase & not Synchronous. 

You CAN have solar charging & Gene charging on Gen Mode. The ATESS UI is horrible, that much is true but there are 2 settings on the Battery charger. Solar only Charging & Grid & Solar simultaneous charging. I know you mentioned you have an EMS, but that product is not serving you if you cannot solar charge when there is Gene power available. 

We have designed our own EMS & to do something like this one must have understood the operation very well. Even if you don't use the EMS look under Sample Calibration, you will find the setting that I'm referring to. It works under any mode, On Grid or Generator mode. Check my screen shot. 

[Steve87]

3 hours ago, macafrican said:

So in essence a hybrid of DC and AC coupled smartgrid

This is possible but in the ATESS architecture, you can only choose the Bypass AC coupled config or the RTF, not both at the same time. The allow you to AC couple on the DC coupled RTF only. The main reason is that to have seemless switching the PCS needs to stay in Off grid mode. But basically you can have an ATS that splits the Gen/ Grid feeding the RTF cabinet. 

To AC couple the Grid tie inverters to your Load side of your PCS you will need a suitable EMS to throttle the Grid tie machines when the Batteries are near full. Not sure if your current EMS allows such. 

In a nutshell the PCS losses it's Bi-directional ability when you use the RTF architecture. 

It sounds like the team that sold or designed your system did not know the ATESS product well at all & now you are paying for that lack of experience with it & it paints a bad name of the ATESS equipment. 

Edited September 28 by Steve87

[BritishRacingGreen]

2 hours ago, macafrican said:

I don’t get that part about rectifier lower output voltage?      

In order to stay safely away from the protection limits of the BMS, in the case where your rectifier  is a dumb charger (no bms comms). 

Remember this : even without bms to charger communication, the bms can still perform its protection functions. 

A conservative charge voltage will not be able to get your pack to 100% SOC.  But you have a use case where you dont need that. The hysteresis levels for  triggering is typically 80%  SOC generator off and say 40% SOC generator on. 

Edited September 28 by BritishRacingGreen

[Steve87]

3 hours ago, macafrican said:

My problem is Atess’s idea of generator integration is a COMPLETE JOKE.  

Its a joke because your pre sales team have made it as such....

[Steve87]

3 hours ago, macafrican said:

My rectifier capacity needs to be close enough to load to be able to at least slow down my rate of battery discharge (if grid down and solar is almost nothing).  

This is exactly what the RTF cabinet can produce...

[macafrican]

53 minutes ago, Steve87 said:

Its a joke because your pre sales team have made it as such....

Well, no.

I had a very precise RFP.

four different large professional installers all ticked the same Yes Can Do boxes in RFP. Probably because Atess manual says explicitly that it can do what was specified.   Atess in writing confirmed the manual had an error in translation. (that can run Gennie and Grid concurrent).

in their model, if Gennie running:

1.  We have to via EMS and PPC keep solar going.    They can if you have their solar charge controller (my solar predated Atess so wasn’t going to change that)

2.  When Gennie running two parameters are key.   One is max generator kW.   So say 345kW for my Scania 420.   The other is charge rate under Gennie which is a STATIC number and you basically have to, unless have a Gennie double your loads, pick a small number.  So because my loads can go 300 : Atess can only charge battery at 45 whether there is 150 solar plus 345 Gennie minus 300 loads or 100 loads or even if I disconnect loads.   I need 300kWh recharge so 7 hours of the Gennie…    That is pathetic in any language.

3.  When trigger Gennie, the two contactors in bypass cabinet kick to open grid and close Gennie.   Then opposite on way out.   The surges in that transfer are so bad I lost Aircon PCB and Chiller power supplies PCB.

there is a lot of the Atess that works great.   Their generator integration is a mistake they will have to fix if they want to play in smartgrid game.

[macafrican]

1 hour ago, Steve87 said:

This is possible but in the ATESS architecture, you can only choose the Bypass AC coupled config or the RTF, not both at the same time. The allow you to AC couple on the DC coupled RTF only. The main reason is that to have seemless switching the PCS needs to stay in Off grid mode. But basically you can have an ATS that splits the Gen/ Grid feeding the RTF cabinet. 

To AC couple the Grid tie inverters to your Load side of your PCS you will need a suitable EMS to throttle the Grid tie machines when the Batteries are near full. Not sure if your current EMS allows such. 

In a nutshell the PCS losses it's Bi-directional ability when you use the RTF architecture. 

It sounds like the team that sold or designed your system did not know the ATESS product well at all & now you are paying for that lack of experience with it & it paints a bad name of the ATESS equipment. 

Atess sent people from China as they had several issues with replacing their PCB and firmware until things worked.

Atess did the job of painting a bad name all on their own!

early warning for people on 500kW bypass cabinet.   They ship with 1250A breakers so if you planned on 630A cables, dig an extra cable trunk now.   Mine also arrived ex factory with the PCB from the smaller one - think 350.   Several daughter board components replaced and several shielded cables replaced.

From EMS and PPC we manage solar so that if grid is off, the solar follows loads from when battery is I think 95%.   When grid is up, we run solar full speed always as we can export to own 11kV transformer.

When their Gennie solution showed itself hopeless, I was going to do what hundreds of sites do every day in SA : run Gennie sync locked onto load busbars.  DeepSea 8610 controller I think it was will manage the sync and govern the Gennie at fixed output.  Atess won’t sign off they can cope…

So hence : I go Gennie : Rectifier : DC combiner with rectifier presenting fixed load to Gennie.    I will therefore have PCS, Isolation Transformer, Bypass Cabinet, Gennie, Rectifier, BESS, Solar and Grid.

another warning : Atess power factor correction is virtually non-existent.    Maybe they assume you have a brand new factory with VSD on everything so almost nothing to manage.   They said they’ll sort PFC and we must rather disable mine.   Under high charge state my measured power factor at 11kV transformer was 0.6!   When your payback is grid capping, it does not help my 90kW cap = 145kVA at council!!!      I solved that by switching my 1MW worth power factor correction back on.   Short debate between my 10 beer keg size capacitors and their two 375ml soda can capacitors

The system generally DOES work.   It keeps within about 8% of my grid demand cap.  I set 90 they might do 95 over half hour measure, so I learnt I set to 86 if I really want 90.    Transfer during loadshed is now virtually not noticeable, which is wonderful.

it would be nice if there was a Watcher (me doing that manually now).    Give it general parameters/goals and then it will figure out that we could run 10kW cap today because solar forecast looks great.   When loadshed becomes an issue again, the watcher operates more carefully.