Arco Zeus regulator - first impressions and a look inside

[IanD]

Just now, dmr said:

 

I think you are correct that Arco intend the N system to have a live feed added to the top brush, this will explain why they suggest putting a fuse in the field wire, I have been puzzling over that. However unless some physical layout of the alternator internals makes a P install difficult I can see no advantage at all in an N install with a live feed to the rotor, and the P arrangement becomes the obvious way to go.

If the P type avoids needing a fuse between field coil and VBAT (because the Zeus makes this connection and includes current limiting) then this is easier -- so long as both field coil terminals are isolated with nothing else connected to them inside the alternator... 😉

[cheesegas]

This all very much sounds like the conversion from internal to external regulation hasn't been done properly. I reckon the internal regulator is still connected to the brushes. For negative regulation, one brush should be connected to +12v and the other via the regulator to negative. Nothing else should be connected to the brushes...

 

The Adverc does not stand alone like the Zeus, it relies on using the internal regulator when it's not pulsing the field up to get 14.4v etc. If you had an Adverc connected previously and simply connected the wire coming out of the alternator to the Zeus, the internal regulator is still connected and will be doing all manner of weird things in parallel with the Zeus! You should have a little pile of bits of potted PCB which you chopped out the alternator to modify it for external regulation.

[dmr]

Just now, IanD said:

If the P type avoids needing a fuse between field coil and VBAT (because the Zeus makes this connection and includes current limiting) then this is easier -- so long as both field coil terminals are isolated with nothing else connected to them inside the alternator... 😉

 

Yes, the field positive supply to the Zeus is fused, and when its not running (transistor off) there is no live feed to the field. I am not sure how much current limiting the Zeus provides and if its transistors can survive a short but that doesn't really matter as its fused.

An N install with a fixed/permanent live feed to the rotor just puts more user added live wires inside the alternator and makes the field wire permanently live (via the rotor resistance) so now has no advantages at all over the P install.

6 minutes ago, cheesegas said:

This all very much sounds like the conversion from internal to external regulation hasn't been done properly. I reckon the internal regulator is still connected to the brushes. For negative regulation, one brush should be connected to +12v and the other via the regulator to negative. Nothing else should be connected to the brushes...

 

The Adverc does not stand alone like the Zeus, it relies on using the internal regulator when it's not pulsing the field up to get 14.4v etc. If you had an Adverc connected previously and simply connected the wire coming out of the alternator to the Zeus, the internal regulator is still connected and will be doing all manner of weird things in parallel with the Zeus! You should have a little pile of bits of potted PCB which you chopped out the alternator to modify it for external regulation.

 

The regulator is suitably disabled, its just a case of cutting a couple of wires.

The Adverc can of course run without an internal regulator (which is what I did) as sometimes the internal regulator voltage is greater than the Adverc low voltage.

So I have done an Adverc style install with the Zeus using the diode trio to supply the field positive. This works fine as long as the duty cycle is not set too low. I now appreciate that Arco intend an N install to have a direct feed to the top brush and feel that a P install is just the better way to go.

[nicknorman]

While you are at it, connect the panel warning light wire that was on D+, to the Zeus "Output 1". Then the alternator warning light will work as normal, plus some fault code blinking.

[dmr]

4 minutes ago, nicknorman said:

While you are at it, connect the panel warning light wire that was on D+, to the Zeus "Output 1". Then the alternator warning light will work as normal, plus some fault code blinking.

Ive already connected the Zeus output1 to the old Adverc light which is functionally identical to the ignition light (same positive feed). This is not currently  working and I can't work out why not, but thats a separate issue. There are software options as to what the light does but I've tried the two obvious choices.

 

 

 

[cheesegas]

1 minute ago, dmr said:

Ive already connected the Zeus output1 to the old Adverc light which is functionally identical to the ignition light (same positive feed). This is not currently  working and I can't work out why not, but thats a separate issue. There are software options as to what the light does but I've tried the two obvious choices.

Have you changed where the other side of the lamp connects to? A standard alternator install will have one side of the lamp connected to positive and the other goes to the alternator which acts as a current sink when the engine is off. However, the Zeus is a current source, not sink; you'll need to connect one side of the lamp to negative and the other to the Zeus.

[nicknorman]

4 minutes ago, cheesegas said:

Have you changed where the other side of the lamp connects to? A standard alternator install will have one side of the lamp connected to positive and the other goes to the alternator which acts as a current sink when the engine is off. However, the Zeus is a current source, not sink; you'll need to connect one side of the lamp to negative and the other to the Zeus.

That is not my reading of the manual. It says

The feature output lamp that can be configured in the app. This
output acts like a switch between your load (light or fan) and the internal ground in the Smart
Regulator. It can switch uo to 48V at 1 Amo.

 

So I read that it's a switch to 0v. Other side of light should be to +12v, probably the IGN switch so it is off then the IGN is off.

9 minutes ago, dmr said:

Ive already connected the Zeus output1 to the old Adverc light which is functionally identical to the ignition light (same positive feed). This is not currently  working and I can't work out why not, but thats a separate issue. There are software options as to what the light does but I've tried the two obvious choices.

 

 

 

Maybe it is not working just because the reg isn't wired up correctly at the moment?

[dmr]

4 minutes ago, cheesegas said:

Have you changed where the other side of the lamp connects to? A standard alternator install will have one side of the lamp connected to positive and the other goes to the alternator which acts as a current sink when the engine is off. However, the Zeus is a current source, not sink; you'll need to connect one side of the lamp to negative and the other to the Zeus.

 

If you are correct then I have done it wrong and this could explain the problem. But if you are correct tthen the Zeus instructions are misleading or just plain wrong..............

"This output acts like a switch between your load (light or fan) and the internal ground in the Smart
Regulator. It can switch uo to 48V at 1 Amo."

 

[nicknorman]

You should just set the "Feature out" to "Warning Lamp" which I think is the default anyway.

[dmr]

5 minutes ago, nicknorman said:

You should just set the "Feature out" to "Warning Lamp" which I think is the default anyway.

 

Yes, and I assume it comes on when the ignition comes on and goes out when the engine starts (alternator starts), just like a standard ignition light, but also flashes error codes. I could try to create an error code tomorrow.

Just maybe my "diode trio" supply is upsetting it but thats unlikely.

The feature-in is purple, the lamp output is purple with a pink trace, maybe there is a harness building error?

[nicknorman]

Just now, dmr said:

 

Yes, and I assume it comes on when the ignition comes on and goes out when the engine starts (alternator starts), just like a standard ignition light, but also flashes error codes. I could try to create an error code tomorrow.

Just maybe my "diode trio" supply is upsetting it but thats unlikely.

The feature-in is purple, the lamp output is purple with a pink trace, maybe there is a harness building error?

Is the light staying on or off?

My reg chip has a few error codes that are reported back to the micro, one being "mechanical failure" ie the thing is not going round (fast enough) and one for overtemp (of the reg chip) and one for electrical failure (which I think is a disconnect in the field circuit). But I just used the "mechanical failure" one to switch on the warning light. The Zeus probably follows a similar logic.

 

If I may say so, just like when one switches from LA to Li, it is hard to shed the LA mindset and switch to the Li mindset, when you switch from a 9 diode internally regulated alternator, to a 6 diode externally regulated alternator (which is what you have done) there is some mindset shedding and switching to be done! Think 6 diode!

[dmr]

The light never comes up (or never came on 😀).......

I've just created an error by disconnecing the alternator thermistor and the light is flashing.

So, it must be my 9 diode install thats upsetting things. This suggests that rather detecting "ignition on, no alternator current" the software is measuring field current flowing and no output current?. I cant decide if this is a good or bad way of doing things.  

Actually its correct....its assuming the field circuit is open circuit so not lighting up, just like a conventional alternator system.

[nicknorman]

11 minutes ago, dmr said:

The light never comes up (or never came on 😀).......

I've just created an error by disconnecing the alternator thermistor and the light is flashing.

So, it must be my 9 diode install thats upsetting things. This suggests that rather detecting "ignition on, no alternator current" the software is measuring field current flowing and no output current?. I cant decide if this is a good or bad way of doing things.  

Actually its correct....its assuming the field circuit is open circuit so not lighting up, just like a conventional alternator system.

It may need the W wire connecting to make everything work properly - otherwise it can’t know how fast the alternator is spinning - or if it’s spinning at all! Can’t remember if you have already connected it?

 

“Actually its correct....its assuming the field circuit is open circuit so not lighting up, just like a conventional alternator system.” NO NO NO!!! You are still thinking 9 diode machine. Wash your mouth out! Think 6 diodes!

Edited February 29 by nicknorman

[dmr]

16 minutes ago, nicknorman said:

It may need the W wire connecting to make everything work properly - otherwise it can’t know how fast the alternator is spinning - or if it’s spinning at all! Can’t remember if you have already connected it?

 

“Actually its correct....its assuming the field circuit is open circuit so not lighting up, just like a conventional alternator system.” NO NO NO!!! You are still thinking 9 diode machine. Wash your mouth out! Think 6 diodes!

 

😀 As I said, the App gives the option of "require Tacho" so I assume it can all work fine without a tach if required, and indeed it does because I initially failed to scroll down far enough to find the tacho option, its in amongst all that CANbus stuff.

I suggest its Arco who are thinking 9 diodes, that are making their warning light a direct replica of an olde worlde alternator 😀

 

Took the TravelPower down to that nice (but expensive) Mr Cox yesterday. He claims that he can also rebuild my cooked Iskra  alternator for less that the price of a new one. Do you think he will give me a discount if I ask him to leave three of the diodes out ? 😀

[TheBiscuits]

1 hour ago, dmr said:

He claims that he can also rebuild my cooked Iskra  alternator for less that the price of a new one

 

Speak to Lancashire Rotating Electrics.  

 

https://lre-ltd.co.uk/

 

 

[cheesegas]

1 hour ago, nicknorman said:

That is not my reading of the manual. It says

The feature output lamp that can be configured in the app. This
output acts like a switch between your load (light or fan) and the internal ground in the Smart
Regulator. It can switch uo to 48V at 1 Amo.

 

So I read that it's a switch to 0v. Other side of light should be to +12v, probably the IGN switch so it is off then the IGN is off.

Maybe it is not working just because the reg isn't wired up correctly at the moment?

Ah you're right, misread it!

[dmr]

Update and a BUG

 

I've got the alternator thermistor on, not looking too horible, but might need to drill and tap the alternator body for a top job.

Have also redone the brush box to give a P type (and 6 diode) configuration. (I can see little point in an N type 6 diode config).

It was good to see that as I changed the setup from N to P the App gave me a serious warning.

This rework (obviously) fixes the excitation issue that I was having.

The warning light still does not function as an ignition light simulation, but flashes on a fault condition so I know its working.

There is a bug where the Zeus occassionally looses control of voltage regulation, I see 15.2volt transients on my 14.8volt target. Could be an issue with Lithiums?????

This happens because the Zeus cycles its duty cycle up and down to control alternator temperature, so switching between a "constant temperature" and "constant voltage" regime. At the transition it sort of takes its eye off the ball and continues to ramp the duty cycle upwards before it gets it back in control and this gives an over-voltage pulse. I am hoping for a good response from Arco. 

[cheesegas]

Just got mine installed but not yet tested. I didn't connect the battery voltage sense lines on purpose - I wanted to see if the Zeus is DVCC compatible, getting battery current/voltage/temp from the BMV in the system. Unfortunately it's not, it absolutely does need its own battery shunt and thermistor in the system which is a real shame; the Wakespeed can take this data over CANbus. Almost every existing install this goes into will have a shunt already in it. I'm unsure of the input impedance of the shunt inputs of the BMV 'brain' and the Zeus; if it's sufficiently high enough, in the 20k ohm range, both can be connected to the same shunt which reduces wiring. I'm going to try it later in the week.

 

Onto the canbus integration side, there's not much info in the manual apart from a one page Victron specific PDF describing how to set up the ve.can ports. Followed the instructions, popped a terminator in the other port in the Cerbo and it popped up in the device list.

 

 

Opened up NodeRed fully expecting to have to create a custom node to pull data off dbus, but firmware v3.14 has added an Alternator input node!

 

 

You can get some useful info from it, but I'd have liked to see alternator temperature in there as well, it's what we all care about with lithium! 

 

Next step is to hook up the battery sense and thermistor cables, then connect the BMV's shunt and see how it behaves...

[dmr]

I think its reasonable that the Zeus needs its own voltage sense. This is a critical bit of info, in fact its the main objective of the Zeus, so it would be risky to rely on some other device providing data over a bus. In addition I suspect that bus issues and delays/slow update rates could play havoc with the control loops (see my post above 😀 ) .

It might be good to have an option to take temperature, and maybe current, from CANbus.

 

I have shared the Victron BMV shunt with the Zeus. Arco do not provide a shunt so I assume that this is their intention, and shunts in series is not a great idea.

There is a post somewhere on a Victron forum saying that sharing shunts is not good, but I feel that even though the signal is small the source impedance of a shunt is so incredibly low that it must be ok.

[cheesegas]

Just now, dmr said:

I think its reasonable that the Zeus needs its own voltage sense. This is a critical bit of info, in fact its the main objective of the Zeus, so it would be risky to rely on some other device providing data over a bus. In addition I suspect that bus issues and delays/slow update rates could play havoc with the control loops (see my post above 😀 ) .

It might be good to have an option to take temperature, and maybe current, from CANbus.

 

I have shared the Victron BMV shunt with the Zeus. Arco do not provide a shunt so I assume that this is their intention, and shunts in series is not a great idea.

There is a post somewhere on a Victron forum saying that sharing shunts is not good, but I feel that even though the signal is small the source impedance of a shunt is so incredibly low that it must be ok.

Good to know the shunt sharing works! They don't recommend a certain shunt or point you to a supplier which is a shame for a product where everything is made as easy as possible to set up, especially when it's such a critical part of the system.

 

The BMV is already providing data to the MPPT and Multiplus across the network, so it would be good if that central data source did everything to avoid errors caused by mismatches. The Zeus has a 'limp' mode where it limits the field; this could be used if it loses voltage data across the network.

[MtB]

Shunts at all, have always intuitively seemed a bodgy idea from a purist electronics POV to me. Regardless of how many. 

 

A bit like you feel about my 2m length of 16mm2 which seems to do much the same job as the Zeus but more clunky, and way cheaper!

 

A different order of magnitude though, I agree!

[dmr]

How quickly does stuff update on a typical CANbus system?.

I suspect the problem is that a sudden increase in engine speed, or more likely a reduction in load current as something turns off, will give an increase in voltage and the Zeus needs to respond to this as fast as possible.

I was trying to look at some transient responses today, and will look again next time I need to run the engine (Tuesday).

[cheesegas]

2 minutes ago, MtB said:

Shunts at all, have always intuitively seemed a bodgy idea from a purist electronics POV to me. Regardless of how many. 

 

A bit like you feel about my 2m length of 16mm2 which seems to do much the same job as the Zeus but more clunky, and way cheaper!

The hybrid long cable thing is an induced voltage drop to reduce current whereas a shunt is a known voltage drop for the purposes of measuring, quite different! The Zeus - and similar regulators - solves problems which hybrid can't, namely reducing alternator load at low RPMs to avoid having the engine idle too slow and run lumpy. And also having a float voltage, so when running a high current load with a full battery you don't send the BMS into a loop. But this isn't a hybrid vs everything else thread...

5 minutes ago, dmr said:

How quickly does stuff update on a typical CANbus system?.

I suspect the problem is that a sudden increase in engine speed, or more likely a reduction in load current as something turns off, will give an increase in voltage and the Zeus needs to respond to this as fast as possible.

I was trying to look at some transient responses today, and will look again next time I need to run the engine (Tuesday).

Almost instantly. It's 1mbit, and on a congested network - which this isn't - you can set the priority of messages to queue them up. 

 

ve.bus is CANbus based, and that's used for time critical functions when linking inverters in parallel; keeping the mains exactly in phase. A lot of modern cars use CANbus for everything now...controlling headlight modules etc etc. 

[MtB]

I still think a shunt is a monumental bodge....

 

 

[dmr]

2 minutes ago, MtB said:

Shunts at all, have always intuitively seemed a bodgy idea from a purist electronics POV to me. Regardless of how many. 

 

A bit like you feel about my 2m length of 16mm2 which seems to do much the same job as the Zeus but more clunky, and way cheaper!

 

A different order of magnitude though, I agree!

 

Yes, shunts are nasty, we spend loads of effort and money on thick cables to get the resistance down, then put a shunt in the circuit. (but adding a length of thin wire as a current limiter is much worse 😀)

My view is that shunts are more acceptable in the positive line, but I am totally outnumberred on this so have just redone things to fall in line.

 

The only and "nicer" alternative is a Hall effect current sensor and currently these are just not accurate enough.

13 minutes ago, cheesegas said:

 A lot of modern cars use CANbus for everything now...controlling headlight modules etc etc. 

 

My new (to me) car reads the speed limit signs, beeps if Im going too fast, and optionaly takes control of driving the car at the correct speed.