Third rail maximum current

[Deepgreen]

I recall from days of yore that, very broadly, 750v applied in the inner areas and 850v further out on some routes, notably the Portsmouth line. That may have been wrong and/or may have changed. From those same times, a 4REP (3,200hp) was not alllowed to operate in multiple with any other traction.

 

[A0wen]

It wouldn't be third-rail if the rolling stock wasn't setting itself on fire...

You'll get all the 3rd rail fanbois on your case making sweeping statements like that about their beloved form of electrification.

 

[A0wen]

I recall from days of yore that, very broadly, 750v applied in the inner areas and 850v further out on some routes, notably the Portsmouth line. That may have been wrong and/or may have changed. From those same times, a 4REP (3,200hp) was not alllowed to operate in multiple with any other traction.

I think the 850v claim was proven to be an urban myth somewhere else on these boards.

REPs not being allowed to run in multiple with other units though I'm sure I read somewhere else (i.e. reputable book not a random Railforums post).

 

[swt_passenger]

I recall from days of yore that, very broadly, 750v applied in the inner areas and 850v further out on some routes, notably the Portsmouth line. That may have been wrong and/or may have changed. From those same times, a 4REP (3,200hp) was not alllowed to operate in multiple with any other traction.

Upgrading the inner areas from 660V to 750V is comparatively recent. 750V was always something of a catch all. SWT inners were still being upgraded to 750V in the early part of CP5, around 2014...

 

[AM9]

I think the 850v claim was proven to be an urban myth somewhere else on these boards. ...

I believe that to be so as well. In the days of the 'REPs', there were also CEPs, CIGs, SUBs and VEPs running around on EE507 motors that were part of a pool of spares, some examples of which went back decades. It's credit to English Electric and Southern Railway that the same parts were truly interchangeable, but with that came the need to ensure that transients on the supply and within the traction electrics didn't create voltage excursions that might exceed the 1930 build standard's insulation limitations.

 

[contrex]

I believe the trains are fitted with 800 amp fuses, so less than that, I presume.

There is a fuse for each shoe; each unit in a train will have at least two in contact with a rail at any one time. If you do the sums, 1 horsepower is 745 watts, which is 1 amp at 750v, near enough, so a 1000 HP unit (typical 4-car slammer) will draw 1000 amps at full chat, more at starting (surge). Modern power-hungry lardbutts take more; an Electrostar can be 1600 HP (i.e. amps). Do the math for coupled units. The substations installed for the Channel Tunnel reinforcements had breakers set to trip at 6,800 amps, I believe. The shoe fuses are made of copper strip, and aren't particularly quick blowing, so they can get quite hot and tolerate short to medium term overcurrents. I remember reading that in the early days of the Bournemouth electrification they had to beef up the substation provision at Waterloo because a REP and some other unit starting at the same time would bring out the breakers, and while that was being carried out, departures had to be staggered. Three minutes comes to mind?

 

[AM9]

There is a fuse for each shoe; each unit in a train will have at least two in contact with a rail at any one time. If you do the sums, 1 horsepower is 745 watts, which is 1 amp at 750v, near enough, so a 1000 HP unit (typical 4-car slammer) will draw 1000 amps at full chat, more at starting (surge). Modern power-hungry lardbutts take more; an Electrostar can be 1600 HP (i.e. amps). Do the math for coupled units. The substations installed for the Channel Tunnel reinforcements had breakers set to trip at 6,800 amps, I believe. The shoe fuses are made of copper strip, and aren't particularly quick blowing, so they can get quite hot and tolerate short to medium term overcurrents. I remember reading that in the early days of the Bournemouth electrification they had to beef up the substation provision at Waterloo because a REP and some other unit starting at the same time would bring out the breakers, and while that was being carried out, departures had to be staggered. Three minutes comes to mind?

Everybody seems to be talking about the 4-REP current draw as it if was way above that of any other 12-car train. The typical formation of a 12-car fast train on the Portsmout Direct was 4-CIG/4-BIG/4-CIG, with a total of 3000HP installed. They were a couple of tons heavier than a 4-REP/4-TC/4-TC train and had just 200 tons less motor power. As far as the 3rd rail and supply chain was concerned, there was little difference between the two. 22 of the 4-BIGs were later modified to become 'Greyhound' units with a weak-field setting allowing a higher balancing speed. Given that this reduced the back-EMF, it may have increased the actual current draw at higher speeds, reducing the difference between the two units even more. *
* I can't remember enough DC machines theory to work out whether the increase in armature current would make for an overall power draw. Maybe anyone here well versed in them might be able to comment.

 

[contrex]

@AM9 - I think the problem was that after the 1967 timetable change, the REPs took considerably more juice out of the 3rd rail than a Bulleid Pacific or BR Class 5 would have done. I expect the effect would have been the same if the SR had provided (e.g.) CIG-BIG-CIG formations (or 6 x 2-HAPs or whatever) for the Bournemouth electrification. That is, the potential peak load to the substation was greater than before, and sometimes the tripping level could be reached.

 

[contrex]

* I can't remember enough DC machines theory to work out whether the increase in armature current would make for an overall power draw. Maybe anyone here well versed in them might be able to comment.

My father was an electrical engineer and would have answered at once; alas, he is no longer alive, and I had to look it up to be sure. In field weakening, you are in the "constant HP range". Torque reduces, and speed increases, by the same factor, so that HP remains the same, equation here or e.g. Donald G. Fink, H. Wayne Beaty "Standard Handbook for Electrical Engineers," 14th Ed., McGraw-Hill, 2000. Equation 8-15 on page 8-21.

 

[edwin_m]

My father was an electrical engineer and would have answered at once; alas, he is no longer alive, and I had to look it up to be sure. In field weakening, you are in the "constant HP range". Torque reduces, and speed increases, by the same factor, so that HP remains the same, equation here or e.g. Donald G. Fink, H. Wayne Beaty "Standard Handbook for Electrical Engineers," 14th Ed., McGraw-Hill, 2000. Equation 8-15 on page 8-21.

There will however be consumption of maximum current over a longer period, because the driver is able to continue accelerating instead of just running at the maximum speed the unmodified train will go.

 

[contrex]

There will however be consumption of maximum current over a longer period, because the driver is able to continue accelerating instead of just running at the maximum speed the unmodified train will go.

I think line speed considerations come into play, don't they? If the line speed is 90, and a train can do 85 without extra field weakening, but can get up to 90 if modified, it won't make any difference between going as fast as the motors allow, and going as fast as the line speed allows, will it? Same horsepower, same current draw, over the stretch of line being considered. I admit I have had to sit and think about this, so I may have missed something.

 

[Taunton]

I think the 850v claim was proven to be an urban myth somewhere else on these boards.

REPs not being allowed to run in multiple with other units though I'm sure I read somewhere else (i.e. reputable book not a random Railforums post).

Well I certainly recall the 850v being documented in a Modern Railways article about the Bournemouth electrification in 1967, for west of Brookwood where it started. Most existing stock could handle it fine, except the older 4-SUB. which however now found it convenient to run to Eastleigh Works for overhaul instead of being dragged; they were not to be allowed in passenger service, and had to have the saloon heaters and lighting switched off.

Each 4-REP motor coach was essentially an electric Class 73 fitted under the frame, same motors etc, so the unit is like a double-headed 73 pair, a common freight arrangement.

 

[AM9]

There will however be consumption of maximum current over a longer period, because the driver is able to continue accelerating instead of just running at the maximum speed the unmodified train will go.

That's similar to my thoughts. If weak field merely extends the rotational range over which the maximum power output is available, that availble power will be usable for longer and in more locations, thereby presenting a greater overall load on the traction supply. Another factor might be that although weak field operation would still be in the 'constant HP range', that presumably is the motor's power output. However, by increasing the armature current, there would be a greater resistive power loss, which in order to maintain the constant HP figure, would involve a greater draw from the supply, which ultimately gets dissipated as heat.

 

[swt_passenger]

Well I certainly recall the 850v being documented in a Modern Railways article about the Bournemouth electrification in 1967, for west of Brookwood where it started. Most existing stock could handle it fine, except the older 4-SUB. which however now found it convenient to run to Eastleigh Works for overhaul instead of being dragged; they were not to be allowed in passenger service, and had to have the saloon heaters and lighting switched off.

Each 4-REP motor coach was essentially an electric Class 73 fitted under the frame, same motors etc, so the unit is like a double-headed 73 pair, a common freight arrangement.

In one of the previous discussions it was claimed that the whole 850V thing originated with a typo in Modern Railways...

 

[Taunton]

In one of the previous discussions it was claimed that the whole 850V thing originated with a typo in Modern Railways...

It wasn't just a potential typo, it was a whole paragraph, including explanation, as above, for what they had to do as a result. And as I think the electrification came on line in July, I don't think it can been an April Fool on editor G Freeman Allen by the engineers either!

Surely someone here knows what the substations put out

 

[swt_passenger]

It wasn't just a potential typo, it was a whole paragraph, including explanation, as above, for what they had to do as a result. And as I think the electrification came on line in July, I don't think it can been an April Fool on editor G Freeman Allen by the engineers either!

Surely someone here knows what the substations put out

I don’t know either way. I’m aware of the possibility of a system being referred to as 750V, but with the substation open circuit voltage being higher, and that voltage reducing under load. Whatever the figure you’d get on a test meter, it always seems to be described as 750V system.

(But when the inner area was usually 660V How was it described?).

 

[edwin_m]

I think line speed considerations come into play, don't they? If the line speed is 90, and a train can do 85 without extra field weakening, but can get up to 90 if modified, it won't make any difference between going as fast as the motors allow, and going as fast as the line speed allows, will it? Same horsepower, same current draw, over the stretch of line being considered. I admit I have had to sit and think about this, so I may have missed something.

Nothing that complicated. It's just that when the unmodified train gets to 85 it will stop drawing much current as the back EMF is opposing most of it. The same will happen to the modified train when it gets to 90, but because it's a higher speed it will take longer to get there. So the substation that feeds the bit of track where trains accelerate from 85 to 90 will be feeding more current to a modified train. This could be significant, as the amount of heat generated in the substation depends on current over time so even if the current is the same it could cause overheating if it is being drawn for longer.

I don’t know either way. I’m aware of the possibility of a system being referred to as 750V, but with the substation open circuit voltage being higher, and that voltage reducing under load. Whatever the figure you’d get on a test meter, it always seems to be described as 750V system.

(But when the inner area was usually 660V How was it described?).

There is a standard for the range of voltages that lineside equipment and trains need to tolerate for a particular nominal voltage, and I'm pretty sure 850V is within the range for a nominal 750V. Assuming the standard was similar in 1967 they could indeed have upped the voltage so that the supply was normally towards the top end of the acceptable range, but better controlled so as not to exceed the maximum. A little bit like how the standards for electrical appliances are based on the European standard 220V but tolerances are wide enough to include the traditional British 240V.

However having the actual voltage set closer to the top of the range does limit the scope for regenerative braking, as a train regenerating will push the voltage up locally, but has to stop regenerating if it goes above the maximum permitted. This wasn't an issue in 1967 but may have become one since.

 

[zn1]

i was taught from day ONE, Juice rails voltage was between 650-750 DC...you never stood on it, you avoided shoe gear when getting on board an emu and you crossed outer 4ft & Juice rail in one step over...or used crossing boards..

it was always banged in to head - respect the the third rail and it will respect you...simple..

it has since the demise of the slam door squadrons been juiced up a bit more to cope with the high demands of modern EMU such leccystar etc and class 92's

 

[contrex]

Well I certainly recall the 850v being documented in a Modern Railways article about the Bournemouth electrification in 1967

That's where the myth started. It is an error.

 

[WAO]

From memory, the tolerance for conductor rail voltage was between 590 and 900V, the higher being possible close to a substation with no current being drawn. When current is drawn, both resistance and transformer regulation cause the voltage to drop. Around 590V, circuit breakers would open and everything would go quiet (except the customers). 750 V is the expected figure but ratings could be taken at c90% of that, as with (c)25kV ac (which also varies quite widely).

There was also something called the Total Conductor Rail Current Index for each train formation. Ordinary SR stock was 2 or 4 per unit but a REP value was 14 not 12, because of motor characteristics. Hence the sparkling performances. The maximum allowed per train was 16.

I imagine that tolerances are now closer. Siemens did struggle with the Dickensian SR power supplies when it won its EMU contract.

Please don't ask for a reference.

WAO

 

[AM9]

From memory, the tolerance for conductor rail voltage was between 590 and 900V, the higher being possible close to a substation with no current being drawn. ...

I suppose that since the introduction of EMUs with regen, the 900V figure also includes any unresolved feedback from such trains.

 

[WAO]

Yes. Solid state power electronics allow regenerated source current from traction motors to be fed back into the conductor rail if there is a sink for it.

What a change from the simplicity of SR camshaft control, EP braking, buckeye couplers and waist level jumpers, where everything connected to everything else!

WAO

 

[Taunton]

I wonder why the UK went for the complexity of the train having to determine whether to regen or not, and having to blend in friction braking.

Elsewhere the train just regens as required, and the substation burns off any occasional surplus current in banks of open lineside resistances, which seems a simpler approach.

 

[edwin_m]

I wonder why the UK went for the complexity of the train having to determine whether to regen or not, and having to blend in friction braking.

Elsewhere the train just regens as required, and the substation burns off any occasional surplus current in banks of open lineside resistances, which seems a simpler approach.

Probably because it would need every substation modifying. In any case it would need friction brake blending for the (rare) occasions when the supply from the substation was lost while braking, or if other trains regenerating nearby had pushed the line voltage up to the maximum permitted.

 

[swt_passenger]

I wonder why the UK went for the complexity of the train having to determine whether to regen or not, and having to blend in friction braking.

Elsewhere the train just regens as required, and the substation burns off any occasional surplus current in banks of open lineside resistances, which seems a simpler approach.

To add to the previous answer the basic difficulty (that would need major modification) is that the transformer/rectifiers in the DC area couldn’t work in reverse, ie as inverters. I don’t know if new or recent kit does, but it’ll probably still be a minority.

 

[AM9]

I wonder why the UK went for the complexity of the train having to determine whether to regen or not, and having to blend in friction braking.

Elsewhere the train just regens as required, and the substation burns off any occasional surplus current in banks of open lineside resistances, which seems a simpler approach.

From memory, the tolerance for conductor rail voltage was between 590 and 900V, the higher being possible close to a substation with no current being drawn. When current is drawn, both resistance and transformer regulation cause the voltage to drop. Around 590V, circuit breakers would open and everything would go quiet (except the customers). 750 V is the expected figure but ratings could be taken at c90% of that, as with (c)25kV ac (which also varies quite widely).

There was also something called the Total Conductor Rail Current Index for each train formation. Ordinary SR stock was 2 or 4 per unit but a REP value was 14 not 12, because of motor characteristics. Hence the sparkling performances. The maximum allowed per train was 16.

I imagine that tolerances are now closer. Siemens did struggle with the Dickensian SR power supplies when it won its EMU contract.

Please don't ask for a reference.

WAO
[/QUOTE]
So don't the Electrostars have rheostatic braking resistors for when the track can't receive regen power?

 

[hwl]

I wonder why the UK went for the complexity of the train having to determine whether to regen or not, and having to blend in friction braking.

Elsewhere the train just regens as required, and the substation burns off any occasional surplus current in banks of open lineside resistances, which seems a simpler approach.

It isn't blending with friction braking, southern region started introducing rheostatic braking (dumping the electricity into resistor banks) nearly 60 years ago so when (Govia) Southern wanted to introduce regen on the electrostars was introduced it was case of now a case of change the software to blend between existing rheostatic and regen back to 3rd rail based on 3rd rail voltage. The blending is electric vs friction depending on speed (friction at jogging pace or less.

If the 3rd rail voltage is too low there is no current returned to 3rd rail for safety reasons.

 

[hwl]

So don't the Electrostars have rheostatic braking resistors for when the track can't receive regen power?

Precisely was used exclusively for the electric braking for the first 8 years before regen back into 3rd rail was authorised.

 

[AM9]

Precisely was used exclusively for the electric braking for the first 8 years before regen back into 3rd rail was authorised.

Thanks, - I assumed that they did because sometimes there was the characteristic smell of cooking resistor banks when they slowed down without brake pad noises.

 

[notadriver]

Everybody seems to be talking about the 4-REP current draw as it if was way above that of any other 12-car train. The typical formation of a 12-car fast train on the Portsmout Direct was 4-CIG/4-BIG/4-CIG, with a total of 3000HP installed. They were a couple of tons heavier than a 4-REP/4-TC/4-TC train and had just 200 tons less motor power. As far as the 3rd rail and supply chain was concerned, there was little difference between the two. 22 of the 4-BIGs were later modified to become 'Greyhound' units with a weak-field setting allowing a higher balancing speed. Given that this reduced the back-EMF, it may have increased the actual current draw at higher speeds, reducing the difference between the two units even more. *
* I can't remember enough DC machines theory to work out whether the increase in armature current would make for an overall power draw. Maybe anyone here well versed in them might be able to comment.

I wish someone could do an actual rail horsepower combination between a 12 car cig and a 12 car rep + 8tc. I believe the motor ratings given were 1 hour ratings - indeed looking at a output curve they produce more than their 1 hour rating during acceleration (peaking at about 40 mph) when in weak field and then output tails off rapidly to below the 1 hour rating especially at higher speeds above 70 mph.
But the balancing speed of a CIG set was in the low 80s on level track. I’ve seen a few logs where they struggled to reach that between Surbiton and Woking. Multiple formations were quicker. 12 car 4 rep + 8TC formations despite only having 200 hp extra on paper had lots of grunt; they could easily sustain and exceed 90 mph on level track and indeed climb the 1/252 to Worting jct at 80 mph compared with unmodified CIGs that could barely make that speed on level track !