Transformer Inrush

[bantam_dave21]

Hi,

I've been reading around the forums and there seems to be a mixed bunch of answers on the topic.

A customer wants to supply some transformers with our diesel generator (which I am sizing up), he's expecting I can and will calculate the inrush currents to ensure in step loading terms our inti can handle it and so we can design any unit so it is large enough to keep voltage and frequency within specified limits for transitory load acceptance.

The customer says he has two tx's in series - and to quote "( 1^st TX – 60 kVA, 0.4/1 kV, 4.5% impedance, 2^nd TX – 80 KVA, 1/0.4 kV, 4% impedance). Transformer inrush current must be considered in your study report. "

Rightly or wrongly I take from the above that they are stepping up to 1kv I'm assuming for volt drop purposes and stepping back down to 400V at the remote end. How do I begin to calculcate the load that I will see at the supply side? What is the calculation I should follow? Do I have enough info to do this, the single line diagram I've been shown shows these supplied from a 200A TP fuse way but the inrush current, if any is my critical info so as to ensure the unit is large enough to accomdate the step load. This is all three phase, 50hz 400 V supply.

Many thanks in advance.

Dave

 

[templdl]

First this isn't a utility supply that you are being concerned with where the supply is very stiff and able to supply instantaneous peaks of current.
Then I nsiderred the generator and what it's current spike capability is.
Then, is in actual transformer designed ft or a step up application.
Consider the inrush tyo be as round 12x the fla, slightly higher for trsnsformer with the LV wdgs wound closest to the core.
I think that you may find that this a pointless Chinese firedril as transformer inrush only becomes an issue with the breaker that supplies the transformer anyway and if a OCPD is required both on the input and output side of the transformer. NEC art 450 would give you some direction on those requirements.

 

[jim dungar]

Is your generator running at rated speed when it experiences the transformer inrush? How will the generator respond to having 15-20X transformer FLA slammed onto it for a second or so? Dealing with transformer inrush is not much different than dealing with motors, except the magnitudes of the currents when a transformer is being used 'backwards'. Having two transformers in series basically extends the inrush time not its magnitude.

 

[GoldDigger]

I would worry about the effects of voltage drop at the generator on other loads it supplies.
The impedance of the generator may well limit the inrush current to the transformer, but it will do that by dropping the output voltage seen by other equipment too.
I would not worry as much about a breaker as I would about a non-self-resetting protective circuit in the generator.

 

[Besoeker]

Hi,

I've been reading around the forums and there seems to be a mixed bunch of answers on the topic.

A customer wants to supply some transformers with our diesel generator (which I am sizing up), he's expecting I can and will calculate the inrush currents to ensure in step loading terms our inti can handle it and so we can design any unit so it is large enough to keep voltage and frequency within specified limits for transitory load acceptance.

The customer says he has two tx's in series - and to quote "( 1^st TX – 60 kVA, 0.4/1 kV, 4.5% impedance, 2^nd TX – 80 KVA, 1/0.4 kV, 4% impedance). Transformer inrush current must be considered in your study report. "

Rightly or wrongly I take from the above that they are stepping up to 1kv I'm assuming for volt drop purposes and stepping back down to 400V at the remote end. How do I begin to calculcate the load that I will see at the supply side? What is the calculation I should follow? Do I have enough info to do this, the single line diagram I've been shown shows these supplied from a 200A TP fuse way but the inrush current, if any is my critical info so as to ensure the unit is large enough to accomdate the step load. This is all three phase, 50hz 400 V supply.

Many thanks in advance.

Dave

First of all, welcome! Which part of Yorkshire if I may ask?
There are a few Brits who post here, myself being one of them.

Transformers come in different styles and different grades of core laminations. The point in the cycle where the switching occurs can have a large bearing on the magnitude. In short, there is no single on size fits all solution. I'll post a link to give you an idea of what's involved but core saturation is a major factor.

The best starting point, in my opinion, would be to contact the manufacturer with the serial numbers and ask their advice.

Anyway, here is the link.

http://www.openelectrical.org/wiki/index.php?title=Transformer_Inrush

 

[templdl]

I apologize as I didn't realize that the OP's intent was to energize the transformer only after the genset was running. I was mistakenly assuming that the transformers were already connected when the genset was started where inrush wouldn't be an issue. I was trying to sort out at what time transformer inrush becomes an issue. If for some reason it has been designed only to connect the transformers after the genset is up to speed then it would be a question as to what current the generator is capably of producing at the time the transformer is connected. If the transformer are the only load then?.
I'm trying to sort out if this is a practical application issue or simply theoretical.
I have found that specs are often put together by cutting and pasting and end up to be not applicable. I simply ask the person who submitted the specs what they are looking for. Too often the person who is supposed to interpret the specs is too embarased to ask the question. I have found that simply asking the question can keep one from chasing ones tail. Thern there is the opportunity to connect with the customer an learn more about their concerns.

 

[bantam_dave21]

Hi everyone and thank you for your responses. I'm based near Ferrybridge with Eggbirough PS out of one window and Ferrybridge the other lol.

To add some more detail, this is a practical application at this point however my client probably hasn't considered the immense over sizing of the generating set required and the financial viability of the oversizing. The normal static load of LV equipment is around 90kva and there is minimal increase for starting loads with the obvious exception of these transformers.

The only theoretical part is the diesel generating set, this will be sized to be large enough to supply all connected load hence the reason for the question really once I can define the load characteristics of the tx's I can begin to ensure we're selecting the correct engine alternator combination. The alternator impedance will effect the current output but probably to the point of stalling.

when I get back at a desk I will digest these replies a little more and piece it together a little more.

thanks again,
Dave

 

[Haji]

You may use a small auxiliary transformer about 5KVA size in the circuit to limit inrush current ( to about 20%) of the main transformers and to reduce size of the diesel generating set also. The auxiliary transformer is to be bypassed after the inrush event is over.

 

[templdl]

Hi everyone and thank you for your responses. I'm based near Ferrybridge with Eggbirough PS out of one window and Ferrybridge the other lol.

To add some more detail, this is a practical application at this point however my client probably hasn't considered the immense over sizing of the generating set required and the financial viability of the oversizing. The normal static load of LV equipment is around 90kva and there is minimal increase for starting loads with the obvious exception of these transformers.

The only theoretical part is the diesel generating set, this will be sized to be large enough to supply all connected load hence the reason for the question really once I can define the load characteristics of the tx's I can begin to ensure we're selecting the correct engine alternator combination. The alternator impedance will effect the current output but probably to the point of stalling.

when I get back at a desk I will digest these replies a little more and piece it together a little more.

thanks again,
Dave

Great Dave, It looks as if you are going to get your arms around this. Knowing the dynamics of the system woud be a benefit.
Dave

 

[gar]

151118-0842 EST

Size the generator somewhat above the actual expected normal peak load.

Put a whopping big inertia flywheel on the generator, or put a current limiting resistor in series with the transformer primary that is shorted after several cycles ( 0.1 second ).

When shutting down open the input to the transformers while there is a large resistive load on the transformers, or gradually reduce input voltage to the transformers towards 0 before removing power input to the transformers.

Various combinations of the above may be useful.

.

 

[steve66]

Generator manufacturers usually have software available to figure voltage drop given specific loads, like motor loads, or HVAC loads, or transformers.

You can either get the software from your rep. (usually free for engineers), or you could contact the rep. and they will probably run the analysis for you.

The software will suggest combinations of generator engine sizes and alternator sizes for a specified voltage drop.

 

[Besoeker]

151118-0842 EST

Put a whopping big inertia flywheel on the generator, or put a current limiting resistor in series with the transformer primary that is shorted after several cycles ( 0.1 second ).

I would think that the prime mover and the alternator would have enough inertia to keep speed from changing significantly for several cycles.

 

[GoldDigger]

I would think that the prime mover and the alternator would have enough inertia to keep speed from changing significantly for several cycles.

I once had a cheap portable generator that used a lawnmower engine and a permanent magnet generator (large magnet that was as large and as heavy as a flywheel. No voltage regulator, you just got what the generator produced at the governed speed.
When I had it located remotely from where I was using the power (noisy!) I would often stop it by shorting the output.
It came to a dead stop in less than three cycles of the four stroke engine (and rocked visible on its base.)

 

[gar]

151119-2023 EST

The maximum continuous power rating and its stall point are close together for an internal combustion engine.

Whereas, an electrical machine can be way overloaded for a short time without damage and will recover as the overload is removed.

Transformer inrush current is a very short time overload, won't damage an alternator, but can easily stall a driving internal combustion engine, and as the overload quickly dissipates the alternator can recover and supply energy as long as its mechanical driver rotates.

A steam engine as the prime mover would have no problem. It would stall and immeadiately recover, but also might have a larger flywheel.

Gasoline engines stall imeadiately when overloaded, and have little excess moving or rotating mass. Gasoline engines are designed with the minimum possible moving mass.

.

 

[Besoeker]

151119-2023 EST

The maximum continuous power rating and its stall point are close together for an internal combustion engine.

Whereas, an electrical machine can be way overloaded for a short time without damage and will recover as the overload is removed.

Transformer inrush current is a very short time overload, won't damage an alternator, but can easily stall a driving internal combustion engine, and as the overload quickly dissipates the alternator can recover and supply energy as long as its mechanical driver rotates.

A steam engine as the prime mover would have no problem. It would stall and immeadiately recover, but also might have a larger flywheel.

Gasoline engines stall imeadiately when overloaded, and have little excess moving or rotating mass. Gasoline engines are designed with the minimum possible moving mass.

.

Depends on application. For most that I have encountered as standby units the IC part dwarfs the alternator.
F1 is a whole different ball game.

 

[bantam_dave21]

Hi everyone,

thanks for your replies so far.

That has been half of the problem, not knowing the exacts of the system and trying to interpret what my client his envisaging doing with this Generator. I now have a slightly better idea but it has took some work to extract from them what it is they're trying to achieve.

The generator is back to the utility supply for tidal generation. The tidal generation units are circa 2km from the supply point so they have decided it is a good idea to supply the offshore units via a 60kva 400v/1KV step up transformer due to the cable length/volt drop/IR losses, this will then subsequently supply 2 x transformers in series at the remote end 30kva 1kv/400v step down.
When on generator supply the generator will also be supporting an initial load step of 45kva@ 0.9 pf (which coincidentally I've just found out is made up of some UPS load, so I need to investigate this now also)

The really comedy element is they have allowed for an 85kva 400V 50hz diesel generator in the belief they will be able to limit the load to less than this, this scheme is partially designed and perhaps even partially built who knows and they have given no consideration to the fact that diesel generators are not the infinite force of mains and all loads placed on them have to be carefully considered and their effect on the generator looked at.
They have no in rush figure for their 400 up to 1KV transformer but the 1KV down to 400V is 6 x inrush current and they have fuse modelling showing circa 350amps inrush per transformer on the HV side???

My head hurts.....

 

[Bugman1400]

The typical inrush for 60Hz xfmrs here in the U.S. is 8-12X FLA. I'm not sure where the 15-20X in a previous post comes from unless that is the rule of thumb for smaller xfmrs. I don't think I've seen any posts that have really addressed the voltage sag. The voltage sag will be directly influenced by the voltage regulator (VR). Does the genset have a newer static VR or an older analog or dynamic VR? The older VRs are significantly slower. More info about the VR must be known (ie type, mfr, response time,etc.). I have also not seen any mention about the load and how susceptible it is to voltage and frequency sag. Some loads (ie resistive loads, lighting, etc.) may not care and will just reflect the sag and then the eventual recovery up to nominal operating frequency and voltage. Other loads may have protective devices on them that will cause the load to trip during frequency or voltage excursions. More info about the load must also be known.

 

[templdl]

The typical inrush for 60Hz xfmrs here in the U.S. is 8-12X FLA. I'm not sure where the 15-20X in a previous post comes from unless that is the rule of thumb for smaller xfmrs. I don't think I've seen any posts that have really addressed the voltage sag. The voltage sag will be directly influenced by the voltage regulator (VR). Does the genset have a newer static VR or an older analog or dynamic VR? The older VRs are significantly slower. More info about the VR must be known (ie type, mfr, response time,etc.). I have also not seen any mention about the load and how susceptible it is to voltage and frequency sag. Some loads (ie resistive loads, lighting, etc.) may not care and will just reflect the sag and then the eventual recovery up to nominal operating frequency and voltage. Other loads may have protective devices on them that will cause the load to trip during frequency or voltage exc
ursions. More info about the load must also be known.

In my study of xfmr inrush values as they related to the pri OCPD instaneous trip calibration higher KVAs have lower inrushes and increase as the KVA gets lower. Expect up to 7kva up to 23x, 45kva 12x, and >150kva to be 8x the FLA. These are typical ball park values to be expected base upon a 150degC rise dry type distribution transformer. 115 and 80degC will be 20-60% higher as well as 'K' factor and the energy efficient designs.
I always recommend that the highest allowable pri OCPD as allowed by the NEC be used to stay as far from a nuisance tripping problem as you can. An additional note to remember other than the small residential breakers expect that the inst trip calibration to be 10x the ration of the breaker with a tolerance of +-20%. It would s likely that they will be calibrated on the high side.

 

[Bugman1400]

In my study of xfmr inrush values as they related to the pri OCPD instaneous trip calibration higher KVAs have lower inrushes and increase as the KVA gets lower. Expect up to 7kva up to 23x, 45kva 12x, and >150kva to be 8x the FLA. These are typical ball park values to be expected base upon a 150degC rise dry type distribution transformer. 115 and 80degC will be 20-60% higher as well as 'K' factor and the energy efficient designs.
I always recommend that the highest allowable pri OCPD as allowed by the NEC be used to stay as far from a nuisance tripping problem as you can. An additional note to remember other than the small residential breakers expect that the inst trip calibration to be 10x the ration of the breaker with a tolerance of +-20%. It would s likely that they will be calibrated on the high side.

Do you have a data sheet or a weblink on the xfmrs that are capable of 150degC rise? That sounds high and I would like to read up on that. Thanks!

 

[gar]

151211-1330 EST

Bugman1400:

The following does not directly address your question, but may be useful information.

https://www.superioressex.com/uploa...s/emcwa-nema_magnet-thermal-class-ratings.pdf

.