EMR Class 810 reliability: potential improvements

[marshlinker]

Moderator note: split from

Class 810 for East Midlands Railway Construction/Introduction Updates

Looks like the new Hitachi AT300's based on the IET platforn for EMR are going to be called the Class 804. These are going to be bi-mode trains and should be capable of running at up to 140mph, even if 125 is all they see in service. We're looking at 33, 5-car long sets, with 23m long coaches...

www.railforums.co.uk

I wonder if down rating the MTU 1600 engine to one of the earlier specs (at least temporarily) would improve reliability?

It seems a pretty heroic aspiration to run a rail engine at 100% of its specification; OK, on diesel it wouldn't be able to match the timings of a 222, but presumably there would be time savings on the electrified section to compensate.

Adding 5 minutes to London to Sheffield timings overall would surely be better than the current situation?

 

[sharpener]

I wonder if down rating the MTU 1600 engine to one of the earlier specs (at least temporarily) would improve reliability?
It seems a pretty heroic aspiration to run a rail engine at 100% of its spec.

You would expect the power rating to be a function of ambient temperature (as with many mechanical and electrical items) so some de-rating in the current heatwave would seem to be very sensible, whether contractually imposed or not.

 

[Railperf]

I wonder if down rating the MTU 1600 engine to one of the earlier specs (at least temporarily) would improve reliability?
It seems a pretty heroic aspiration to run a rail engine at 100% of its spec.
Ok, on diesel it wouldn't be able to match the timings of a 222 but presumably there would be time savings on the electrified section to compensate.
Adding 5 mins to London to Sheffield timings overall would surely be better than the current situation?

The performance being developed at the rail indicated by the acceleration achieved from each station stop suggests the Class 810 engines are detuned to much lower levels than the maximum quoted outputs.

At maximum diesel engine output on diesel they should be able to deliver up to 10% better performance than a Class 222 as they do on electric. But for some reason they are behaving like a Meridian that has lost an engine. And on many days there are sets performing as if they are running on three rather than four engines. I've been on a set that was so slow, it felt it was running on only two engines - confirmed by the driver.

 

[marshlinker]

The performance being developed at the rail indicated by the acceleration achieved from each station stop suggests the Class 810 engines are detuned to much lower levels than the maximum quoted outputs.

At maximum diesel engine output on diesel they should be able to deliver up to 10% better performance than a Class 222 as they do on electric. But for some reason they are behaving like a Meridian that has lost an engine. And on many days there are sets performing as if they are running on three rather than four engines. I've been on a set that was so slow, it felt it was running on only two engines - confirmed by the driver.

According to the specs sheets a 5 car 810 has:
4 X Mtu 12 V 1600 R91 rated at 735 KW =2940
and an equivalent 5 car 222 has:
5 X Cummins QSK 19 rated at 555 KW = 2795
Assuming the trains are approximately the same weight the theoretical power to weight advantage of the 810 would be max of 5%
However the MTU is a tier V engine whilst the Cummins was a tier 2/3 engine. I'm not sure how the auxiliaries compare.
I think if you are expecting better performance from the 810 on diesel you are likely going to be disappointed.
Matching the performance is likely to be a real stretch.

 

[Prime586]

You would expect the power rating to be a function of ambient temperature (as with many mechanical and electrical items) so some de-rating in the current heatwave would seem to be very sensible, whether contractually imposed or not.

The MTU powerpacks use direct injection controlled via an ECU. All modern ECUs have charge air temperature and static air pressure sensors, the data from which are used as inputs to the ECU's fuelling and boost pressure control maps. The ECU automatically reduces output power via limiting boost and fuel as ambient air temperature increases.

== Doublepost prevention - post automatically merged: 30 Jul 2026 ==

However the MTU is a tier V engine whilst the Cummins was a tier 2/3 engine. I'm not sure how the auxiliaries compare.

The 22xs date from before 2006 when the first EU emissions regs that applied to rail (Stage IIIa) were introduced. The Cummins engines were however developed to meet the Euro II Road emissions regs at the time, which required no external particulate filtering or NOx control equipment.

The 810's MTU powerpacks comply to EU Stage V and are fitted with DPFs and Selective Catalytic Reduction (Urea/AdBlue injection) to reduce NOx emissions.

 

[222001]

According to the specs sheets a 5 car 810 has:
4 X Mtu 12 V 1600 R91 rated at 735 KW =2940
and an equivalent 5 car 222 has:
5 X Cummins QSK 19 rated at 555 KW = 2795
Assuming the trains are approximately the same weight the theoretical power to weight advantage of the 810 would be max of 5%
However the MTU is a tier V engine whilst the Cummins was a tier 2/3 engine. I'm not sure how the auxiliaries compare.
I think if you are expecting better performance from the 810 on diesel you are likely going to be disappointed.
Matching the performance is likely to be a real stretch.

While the engines have more power in an 810, I believe the the combined motor power is actually less than a 222, which might explain the performance difference?

 

[Railperf]

While the engines have more power in an 810, I believe the the combined motor power is actually less than a 222, which might explain the performance difference?

No. The 810's motors produce faster acceleration in electric mode than a healthy Class 222 on all five engines. So there cannot be a deficit in power output of the eelctric motors

 

[222001]

No. The 810's motors produce faster acceleration in electric mode than a healthy Class 222 on all five engines. So there cannot be a deficit in power output of the eelctric motors

The engines must be limited in their output then, otherwise it doesn't make any sense.

 

[sharpener]

According to the specs sheets a 5 car 810 has:
4 X Mtu 12 V 1600 R91 rated at 735 KW =2940

The MTU powerpacks use direct injection controlled via an ECU. All modern ECUs have charge air temperature and static air pressure sensors, the data from which are used as inputs to the ECU's fuelling and boost pressure control maps. The ECU automatically reduces output power via limiting boost and fuel as ambient air temperature increases.

Thanks. So does the engine mapping in the ECU give a minimum of 735 kW even at high ambient temps or is it the case that this figure applies only in more temperate (shoulder season) conditions and is derated ATM?

 

[Railperf]

I understand 735kw is the absolute maximum the engines are capable of. However, Hitachi / MTU seem to have tuned the engines outputs to meet a specific acceleration / reliability requirement - which is not publicly stated.

A similar situation exists with the original 80x fleets - particularly GWR, TPE and Hull Trains. At 100% power demand from the controller, GWR drivers report the trains TMS screen reporting 80% maximum engine output. THis comment from another forum from such a driver:

"I put the engine screen up on the TMS and drive it like I stole it, until the engines get a bit warm and then I ease off a bit, although most of them are now set to around 80% maximum power instead of the 90-95% they were previously so they dont tend to get hot now, but we cant keep time either but hey ho"

 

[Prime586]

Thanks. So does the engine mapping in the ECU give a minimum of 735 kW even at high ambient temps or is it the case that this figure applies only in more temperate (shoulder season) conditions and is derated ATM?

That will be a maximum figure. The spec sheet quotes the output at "Ambient conditions for railcar: -25°C to 40°C", though once ambient starts getting up past say 25 degrees it will start having an effect (if you have a direct injection turbodiesel car you can notice this in summer - they always feel a bit 'strangled' on hot air days compared to autumn or spring) . As @Railperf says above, the operators may choose to further derate the output (the ECUs can store a number of tuning maps which may be selected by the maintenance staff) for fuel efficiency, reliability or to extend service intervals.

 

[Wyrleybart]

According to the specs sheets a 5 car 810 has:
4 X Mtu 12 V 1600 R91 rated at 735 KW =2940
and an equivalent 5 car 222 has:
5 X Cummins QSK 19 rated at 555 KW = 2795
Assuming the trains are approximately the same weight the theoretical power to weight advantage of the 810 would be max of 5%
However the MTU is a tier V engine whilst the Cummins was a tier 2/3 engine. I'm not sure how the auxiliaries compare.
I think if you are expecting better performance from the 810 on diesel you are likely going to be disappointed.
Matching the performance is likely to be a real stretch.

I am not an engineer or techie so apologies if I have misinterpreted it, but I think there is a subtle difference here.
The 5 car 222 had five Cummins QSK19s each spinning a traction alternator which delivers via the Agate electrical system to a traction motor at each end of each car driving the inner axles. So assuming the 222s have the same traction motors as a 220/221 then a five car 222 will have ten 375kW traction motors - powering every other axle on the train.

I believe the 810 had four MTU engines spinning four traction alternators which feed four traction motors under each of cars 2 and four - 8 motors in total. So the 810 has 8/20 powered axles whereas the 222 has ten. Google suggests that the traction motors are rated at 290kW.

So assuming Google is correct (?) the class 222 has 3750kW and the 810 has 2900kW in terms of traction motor rating.

What is not known is if, say car 2 has the MTU isolated, does the traction power reduce by 25% ? Or can three MTUs make up any traction shortfall - obviously anticipating that some engine output from the remainng MTUs will be required to supply hotel power to the isolated engined car.

 

[sharpener]

"I put the engine screen up on the TMS and drive it like I stole it, until the engines get a bit warm and then I ease off a bit, although most of them are now set to around 80% maximum power instead of the 90-95% they were previously so they dont tend to get hot now, but we cant keep time either but hey ho"

That will be a maximum figure. The spec sheet quotes the output at "Ambient conditions for railcar: -25°C to 40°C". As @Railperf says above, the operators may choose to further derate the output (the ECUs can store a number of tuning maps which may be selected by the maintenance staff) for fuel efficiency, reliability or to extend service intervals.

Sounds like in practice the 736kW @40C in the spec sheet isn't achievable in a hot summer then.

I note this wikipedia article repeats a Modern Railways comment that they are prone to overheat. And it also says the GWR units (which have a slightly different 700kW version) have reverted to their 100% rating, but that is not consistent with the driver's experience.

Quarts and pint pots spring to mind. Also the long history of the Paxman Valentas.

 

[Railperf]

I am not an engineer or techie so apologies if I have misinterpreted it, but I think there is a subtle difference here.
The 5 car 222 had five Cummins QSK19s each spinning a traction alternator which delivers via the Agate electrical system to a traction motor at each end of each car driving the inner axles. So assuming the 222s have the same traction motors as a 220/221 then a five car 222 will have ten 375kW traction motors - powering every other axle on the train.

I believe the 810 had four MTU engines spinning four traction alternators which feed four traction motors under each of cars 2 and four - 8 motors in total. So the 810 has 8/20 powered axles whereas the 222 has ten. Google suggests that the traction motors are rated at 290kW.

So assuming Google is correct (?) the class 222 has 3750kW and the 810 has 2900kW in terms of traction motor rating.

What is not known is if, say car 2 has the MTU isolated, does the traction power reduce by 25% ? Or can three MTUs make up any traction shortfall - obviously anticipating that some engine output from the remainng MTUs will be required to supply hotel power to the isolated engined car.

There is a clear reduction in acceleration based on how many engines are running. So if one of four engines fails, then the acceleration rate matches what would be expected of a 3-engined set rather than a slight reduction due to other engines increasing in power level.
And that seems to reflect what happes in GWR land where a 9-car set on three out of five engines is described as painfully slow. I think the ability for engines to uprate power in case of one failing may have been possible at the original power setting of 750hp - because each of the remaining engines (only two on a five car set) had more extra power in reserve.
On GWR it was said engines used to shut down once the coolant got above 115c, but now they now seem to derate / idle at about 110c

== Doublepost prevention - post automatically merged: 30 Jul 2026 ==

The 5 car 222 had five Cummins QSK19s each spinning a traction alternator which delivers via the Agate electrical system to a traction motor at each end of each car driving the inner axles. So assuming the 222s have the same traction motors as a 220/221 then a five car 222 will have ten 375kW traction motors - powering every other axle on the train.

So assuming Google is correct (?) the class 222 has 3750kW ( in terms of traction motor rating.

What is not known is if, say car 2 has the MTU isolated, does the traction power reduce by 25% ? Or can three MTUs make up any traction shortfall - obviously anticipating that some engine output from the remainng MTUs will be required to supply hotel power to the isolated engined car.

Not sure about your Class 222 power figures, but acceleration measurements based on a computer modeller suggest the 5-car Class 222's are delivering 565hp (421kW) at the rail from an engine setting of 700hp. Hence a unit output at the rail of 2825hp (2107kW). Which is appx 80% of engine power delivered to the rail with 20% used for powering control systems, air con, lighting and other ancilaries.

 

[sharpener]

So assuming Google is correct (?) the class 222 has 3750kW and the 810 has 2900kW in terms of traction motor rating.

What is not known is if, say car 2 has the MTU isolated, does the traction power reduce by 25% ? Or can three MTUs make up any traction shortfall - obviously anticipating that some engine output from the remainng MTUs will be required to supply hotel power to the isolated engined car.

ETA bit of an overlap with @Railperf's post.

I think not, this article says there are 8 x 290kW traction motors on the 810, under cars 2 and 4.

So there must be inter-car connectors for the engines in cars 1 and 5, I would expect it to be a bus capable of handling any combination of power sources incl the xformer under car 3.

And 2 x 290kW is only 580kW which is about 80% of the quoted 736 kW for the R91 version of the MTU leaving a lot for engine derating and ancillary loads. But not enough to power 8x290kW with an engine out. What we don't know is the short-term rating of the traction motors, they may be able to deliver a bit more for a few minutes.

ETA Taking @Railperf's figures if the hotel loads are 20% of engine/generator output I don't see the point of fitting 2 x 375kW motors per car if there is only 80% of 700kW i.e. 560kW actually available.

 

[Railperf]

ETA bit of an overlap with @Railperf's post.

I think not, this article says there are 8 x 290kW traction motors on the 810, under cars 2 and 4.

So there must be inter-car connectors for the engines in cars 1 and 5, I would expect it to be a bus capable of handling any combination of power sources incl the xformer under car 3.

And 2 x 290kW is only 580kW which is about 80% of the quoted 736 kW for the R91 version of the MTU leaving a lot for engine derating and ancillary loads. But not enough to power 8x290kW with an engine out. What we don't know is the short-term rating of the traction motors, they may be able to deliver a bit more for a few minutes.

ETA Taking @Railperf's figures if the hotel loads are 20% of engine/generator output I don't see the point of fitting 2 x 375kW motors per car if there is only 80% of 700kW i.e. 560kW actually available.

With only eight motored axles, the 810 needed fewer, more powerful motors than other Class 80x and what is fitted may be the closest match from the Hitachi Rail's motor catalogue. The motors can only be fed with the current available from OLE through the transformer or diesel engines alternator.

We know the Class 222's offer 11.3 hp/t at the rail, and the 810 on electric mode is closer to 12.4hp / ton - which seems to be reflected by the 810 electric acceleration being faster than a 222.
What is happening in diesel mode is anyone's guess.

As mentioned above the maximum installed diesel power appears to be more than enough to deliver equivalent power to the traction motors after other power consumption. But the fact that the acceleration levels are so much slower suggests modest one of the following - a modest diesel power setting for reliability, or train systems that consume far more than the 20% consumed by a Meridian train systems. Or a cross between the two.
But based on the tinkering with GWR diesel engine outputs, i'm more inclined to go with the engine outputs being tuned downwards for reliability.

 

[Amos390]

Another question, if the units that get sent to Long Marston go to EMR with less faults and issues (maybe if an issue is seen at Marston and fixed), what are the odds that the more troublesome units that are currently with EMR will be swapped out for these functional ones allowing for no decrease in fleet size, but (hopefully) more reliable sets?

 

[liamf656]

Another question, if the units that get sent to Long Marston go to EMR with less faults and issues (maybe if an issue is seen at Marston and fixed), what are the odds that the more troublesome units that are currently with EMR will be swapped out for these functional ones allowing for no decrease in fleet size, but (hopefully) more reliable sets?

You're forgetting that the 222s are supposed to be going off lease before the end of 2026, so it would be a huge reduction in fleet size

 

[Amos390]

You're forgetting that the 222s are supposed to be going off lease before the end of 2026, so it would be a huge reduction in fleet size

Oh yeah, good point, maybe for every 2 in they take 1 out or something. It’ll be a short-term pain for long-term gain scenario, I’m sure they’d rather take the hit on fleet size now but get all the issues fixed rather than sit with the same issues on the same units over and over for months or years to come.

 

[Railperf]

Some Interesting points made by Richard Bowker on Green signals on what the options might be to remedy the situation : includes separating maintenance of 810 and 222 fleets and sit down with Hitachi to work out a reasonable plan for remedying the faults and allowing additional time for this despite any optimism that may come from Hitachi suggesting the issues can be fixed quickly.

 

[800001]

Another question, if the units that get sent to Long Marston go to EMR with less faults and issues (maybe if an issue is seen at Marston and fixed), what are the odds that the more troublesome units that are currently with EMR will be swapped out for these functional ones allowing for no decrease in fleet size, but (hopefully) more reliable sets?

As far as am aware it's one unit going for testing.
I've not heard about anymore units due to visit.

 

[Wyrleybart]

Some Interesting points made by Richard Bowker on Green signals on what the options might be to remedy the situation : includes separating maintenance of 810 and 222 fleets and sit down with Hitachi to work out a reasonable plan for remedying the faults and allowing additional time for this despite any optimism that may come from Hitachi suggesting the issues can be fixed quickly.

I would have thought the 222s would be in a different part of Etches Park to the 810s for maintenance, but maybe not.
The point is that 222s can only reduce in number as 810s increase, so it really is a dynamic balance - this assuming each 810 now becomes a one for one with a 222, and multi working reestablishes as the light at the end of the tunnel. Have EMR ever released an intended final set of diagrams - in other words which services are intended to be doubles and split etc

 

[Clarence Yard]

With only eight motored axles, the 810 needed fewer, more powerful motors than other Class 80x and what is fitted may be the closest match from the Hitachi Rail's motor catalogue. The motors can only be fed with the current available from OLE through the transformer or diesel engines alternator.

We know the Class 222's offer 11.3 hp/t at the rail, and the 810 on electric mode is closer to 12.4hp / ton - which seems to be reflected by the 810 electric acceleration being faster than a 222.
What is happening in diesel mode is anyone's guess.

As mentioned above the maximum installed diesel power appears to be more than enough to deliver equivalent power to the traction motors after other power consumption. But the fact that the acceleration levels are so much slower suggests modest one of the following - a modest diesel power setting for reliability, or train systems that consume far more than the 20% consumed by a Meridian train systems. Or a cross between the two.
But based on the tinkering with GWR diesel engine outputs, i'm more inclined to go with the engine outputs being tuned downwards for reliability.

Hitachi will always set the software for a power curve nearer the minimum power to meet the end to end journey times for the operation rather than the absolute maximum that can be attained by the engine. They do this for engine economy reasons.

Engines on all 8xx units have always been governed in this way, to a greater or lesser degree. Getting Hitachi to give you more is a major task, as FG/GWR found out.

 

[Merle Haggard]

I would have thought the 222s would be in a different part of Etches Park to the 810s for maintenance, but maybe not.
The point is that 222s can only reduce in number as 810s increase, so it really is a dynamic balance - this assuming each 810 now becomes a one for one with a 222, and multi working reestablishes as the light at the end of the tunnel. Have EMR ever released an intended final set of diagrams - in other words which services are intended to be doubles and split etc

Well the newly built shed nearest the line at the Derby station end of the site did have a big sign on it that said 'Hitachi' and 'Home of the Aurora' - or similar (together with the usual meaningless mission statement) The other stock seems to be maintained in the shed behind that, or one further towards Trent.
I say 'did' because I'm not sure the sign's still there.

 

[sharpener]

Hitachi will always set the software for a power curve nearer the minimum power to meet the end to end journey times for the operation rather than the absolute maximum that can be attained by the engine. They do this for engine economy reasons.

Engines on all 8xx units have always been governed in this way, to a greater or lesser degree. Getting Hitachi to give you more is a major task, as FG/GWR found out.

Well the GWR driver quoted in post #7,364 upthread says they won't then keep to the t/table. And my limited amount of experience of PAD - TOT is a lot of lost time in good weather for no very obvious cause, dry track and TSRs on the Berks and Hants section can't be the only reason.

IMO the spec vs t/table should allow the IETs to recover from minor delays not suffer them to propagate, though I see from another thread that is not a preferred view any more. OH travelled to Aberdeen on 1S11 yesterday, it was held up between Stevenage and Hitchin by 6 mins (why?) and arrived 16 down which these days I suppose counts as a good result.

 

[irish_rail]

I understand 735kw is the absolute maximum the engines are capable of. However, Hitachi / MTU seem to have tuned the engines outputs to meet a specific acceleration / reliability requirement - which is not publicly stated.

A similar situation exists with the original 80x fleets - particularly GWR, TPE and Hull Trains. At 100% power demand from the controller, GWR drivers report the trains TMS screen reporting 80% maximum engine output. THis comment from another forum from such a driver:

"I put the engine screen up on the TMS and drive it like I stole it, until the engines get a bit warm and then I ease off a bit, although most of them are now set to around 80% maximum power instead of the 90-95% they were previously so they dont tend to get hot now, but we cant keep time either but hey ho"

This isn't quite true, as the engines will generally give c90 percent power at low speed which as speed increases reduces to c80 percent. But for the first 20 or 30 mph or so (on GWR at least) the 80xs will get 90 percent power from the engines.
The above is based on personal experience as a driver, not sure what the textbook actually says.

 

[Teapot42]

I would have thought the 222s would be in a different part of Etches Park to the 810s for maintenance, but maybe not.
The point is that 222s can only reduce in number as 810s increase, so it really is a dynamic balance - this assuming each 810 now becomes a one for one with a 222, and multi working reestablishes as the light at the end of the tunnel. Have EMR ever released an intended final set of diagrams - in other words which services are intended to be doubles and split etc

Presumably having had several 810s delivered which aren't fit to enter service isn't helping matters in terms of depot space.

Have the 810s been cleared to work in multiple yet? I saw testing was progressing but I've not seen any timescales or outcomes. The temporary timetable not withstanding, but I believe they can't use that many more in service until they can work in multiple - without altering diagrams. It was hinted elsewhere that a significant recast of services will happen once the 810s are all in service, and with more units than there were 222s, I'd imagine more services will run as 10 cars.

Was there any intention to extend the network, or have the likes of Leeds, Scarborough etc been permanently discarded?

It's been suggested that the 222s are due to start going off lease pretty imminently, and at a fairly rapid rate. Is there chapter and verse on this anywhere, rather than a twitter post from someone who may or may not actually have the gen?

 

[JonathanH]

Was there any intention to extend the network, or have the likes of Leeds, Scarborough etc been permanently discarded?

Permanently discarded. The 810 fleet is sized to operate the 2tph to each of Sheffield / Nottingham service.

 

[800001]

Presumably having had several 810s delivered which aren't fit to enter service isn't helping matters in terms of depot space.

Have the 810s been cleared to work in multiple yet? I saw testing was progressing but I've not seen any timescales or outcomes. The temporary timetable not withstanding, but I believe they can't use that many more in service until they can work in multiple - without altering diagrams. It was hinted elsewhere that a significant recast of services will happen once the 810s are all in service, and with more units than there were 222s, I'd imagine more services will run as 10 cars.

Was there any intention to extend the network, or have the likes of Leeds, Scarborough etc been permanently discarded?

It's been suggested that the 222s are due to start going off lease pretty imminently, and at a fairly rapid rate. Is there chapter and verse on this anywhere, rather than a twitter post from someone who may or may not actually have the gen?

Yes they are now passed for 10 car operation.

 

[Bald Rick]

Adding 5 mins to London to Sheffield timings overall would surely be better than the current situation?

That would require a complete timetable rewrite.