Recently was asked to look at a faulty Victron multiplus 3Kva 48V inverter. The symptom was that on switching inverter switch on, when unit attempts to start inverter (with or without load attached) unit emits load buzz noise, then only the overload led comes on and there is no output.

Problem turned out to be one of the two torroid transformers 220V winding was internally shorted. The torroids consist of a 220V winding and two presumably 48V windings which are simply run in paralell. Interestingly there are 2 x torroid transformers in the Victron units and each torroid has separate "H" bridge FET bank to drive the primary windings and the 220V secondary windings are simply paralelled to supply the load.

 

To my limited knowledge all the victron inverters share the same basic design so this info may be useful to you guys out there when you come accross this symptom.

 

Interestingly the user claims the problem occurred without warning. Also interesting is that the transformers have built in temp sense devices in the windings, so its hard to imagine how the transformer windings could have got 'cooked' without the unit having shut down. Takes a fair amount of effort and ingenuity to fry windings on large torroids.Uusally the FETS go for a smaoke break first!

6 minutes ago, ralph said:

To my limited knowledge all the victron inverters share the same basic design so this info may be useful to you guys out there when you come accross this symptom.

The smaller models have only one toroid. Other than that, yes, topology is very similar. Even some phoenix models have essentially the same bits and pieces as the equivalent Multiplus (they just lack the charger and transfer switch).

The newer models are a bit more exciting in terms of microprocessor stuff... :-)

Do you know where one can find any circuit diagrams or tech info on the victron stuff? Seems like theres nothing out there.I must say the victron stuff is impressive in terms of build quality and resilience Ive installed a number of larger quattro's in the field 10 and 15Kva models and they are very resilient to abusive loads.

However they are not so efficient in terms of DC conversion losses (round 80%) in my large off grid solar systems (no fault of thiers, just the limitation of LF transfromers) - I am beginning to think of using HF inverters again for the inverter part of my offgrid sytems simply not to lose the additional 10-15%. energy in the system (difference between efficiency of the LF and HF inverters). Adding 15% to the battery capacity is no joke in the larger systems  - one begins to think hard about losing 15% of your power for the sake of a more bullet proof inverter topology.

15 hours ago, ralph said:

Do you know where one can find any circuit diagrams or tech info on the victron stuff? Seems like theres nothing out there.I must say the victron stuff is impressive in terms of build quality and resilience Ive installed a number of larger quattro's in the field 10 and 15Kva models and they are very resilient to abusive loads.

They want you to send it to an approved repairer. If anything on the control board is blown that is your best bet in any case.

15 hours ago, ralph said:

However they are not so efficient in terms of DC conversion losses (round 80%)

I get 85% to 98% depending on load. When running flat out my 3KVA is maybe around 91%, it improves slightly if you keep it low enough so the fans stay off (eg right now I have around 700w total load, DC vs AC power shows 98% efficiency), but because the no-load draw is around 35W (for a 3kva) it gets bad at low loads.

On off-grid systems that no-load draw over night does add up quite a bit. Especially when you start building 3-phase or parallel models. You can easily lose 7% SoC overnight on a 10kwh battery (actual question I got from someone).

15 hours ago, ralph said:

limitation of LF transfromers

How old are those? The new models (at least the last 6 years or so if not longer) all use a high frequency design. Don't be fooled by the large toroids, if it was low frequency those things would be even bigger.

HI Sorry, my bad for using wrong terminology. The victron multi I'm looking at drives the (large torroid) transformers with HF PWM sine wave at low voltage. The HF designs I was referring to are similar in so far as HF pwm but no large torroids/ large copper transformers involved, for eg RCT/ voltronics type units.

Ref my  comments on efficiency, I refer to the Victron info below.

https://www.victronenergy.com/upload/documents/Output-rating-operating-temperature-and-efficiency.pdf

This indicates 86% efficiency for a multi 3K at close to full load. I shall assume similar stats for larger of their inverters. I think I dont quite get as good figures by a few percent in real life. There are probably a few external factors (eg wiring losses) that will add a few percent losses.

Whlie we are discussing this, I am realising I havent measured up, and this prompts me to go make some detailed measurements on my site when next there, then I can make a precision comment.

Do you know of a victron approved repair facility in SA that fixes boards? or is it a case of module replacement?

 

14 hours ago, ralph said:

This indicates 86% efficiency for a multi 3K at close to full load

That sounds about right. The effiency goes up with load up to a certain point, then it drops back a bit towards the end as the heat losses and cooling costs starts to add up. But that is purely measured in terms of power on the AC side divided by power in on the DC side. There is also an overall efficiency in which the self-discharge of the batteries and the quiescent draw of the inverter plays a large role, the bit you lose overnight that does no useful work, and that does get quite a lot especially when you have multiple units. For example, a three phase setup of 3kva Multis, that can draw easily 100W, which over night can easily amount to 1kwh.

14 hours ago, ralph said:

no large torroids/ large copper transformers involved

Yup, other makers are quite fond of using the smaller E-core ferrites. It is much cheaper of course, which is the main reason they do it (they can get away with it). This is where you get into transformer theory, air gaps and core losses and EMI. Let's just say the E-core is not necessarily a selling point of the other inverters...

The old guys here know me, I always scoff at the salesman's "transformerless" talk. They are fond of calling these inverters with the E-core "transformerless" and then talk as if it is a good thing :-)