37/4 - why the ETH limit?

[Magdalia]

I do think it's a shame that they were always so unhappy about 33s being away from the Southern and didn't shift some of them to use on the kind of general-service duties the 31/4s ended up getting used on. The performance difference seemed out of all proportion to the difference on paper in engine output.

But that's exactly what happened, with class 33 operation extended to Milford Haven and Manchester.

It is quite difficult to do a direct comparison of class 31/4 and class 33 because they rarely worked the same trains, Bristol-Cardiff is one route where someone might have some decent data, rather than basing assertions on anecdotes.

Class 33 ETH was rudimentary, tapped off the main generator, the technology available at the time, and ETH index was 48. It was only designed for Mark Is but could be used on newer stock up to Mark IId, broadly the same restrictions as a Deltic. So class 33s were never called upon to work long trains of aircon stock, apart from odd occasions on the Poole-Newcastle. And in the period when class 33s worked the Waterloo-Exeters the trains were booked load 9 in summer but reduced to load 8 in winter so that timings could be maintained with the ETH on. They were not an ETH loco with operational flexibility like the class 31/4.

What turned out to be a big advantage for the class 33 was that it was built ETH only so never had to carry around a boiler and a tankful of water.

 

[CW2]

I do think it's a shame that they were always so unhappy about 33s being away from the Southern and didn't shift some of them to use on the kind of general-service duties the 31/4s ended up getting used on. The performance difference seemed out of all proportion to the difference on paper in engine output.

But that's exactly what happened, with class 33 operation extended to Milford Haven and Manchester.

Indeed. The use of 33s on the Manchester - South Wales services was generally pretty successful. It was possibly the hardest work they did in their careers, racking up relatively high mileages at speeds and power outputs at the top end of their range. There were relatively few failures (compared with when class 25s used to work on the Marches, for example). I believe the transfer of class 33s to Scotland was looked at more than once, but their Route Availability / maximum axle load ruled them out in favour of converting 37/0s to 37/4s.

 

[Pigeon]

But that's exactly what happened, with class 33 operation extended to Milford Haven and Manchester.

Yes, indeed it was 33s on the Marches that first impressed me with themselves. I was thinking of that operation as being an exception, and wishing there had been more like that.

Class 33 ETH was rudimentary, tapped off the main generator, the technology available at the time

You sure? This lot seem to reckon they had a separate machine: http://sremg.org.uk/diesel/class33.shtml

and ETH index was 48. It was only designed for Mark Is but could be used on newer stock up to Mark IId, broadly the same restrictions as a Deltic. So class 33s were never called upon to work long trains of aircon stock, apart from odd occasions on the Poole-Newcastle. And in the period when class 33s worked the Waterloo-Exeters the trains were booked load 9 in summer but reduced to load 8 in winter so that timings could be maintained with the ETH on. They were not an ETH loco with operational flexibility like the class 31/4.

I agree the ETH index of the 31/4s was remarkably high, but beyond their original purpose, there wasn't much call for it (in terms of booked duties, as opposed to standing in for something bigger that had gone phut). They used to trundle around the place with about 4 Mk 1s on the back, which is well within the ETH capability of a 33. We ended up seeing 33s being withdrawn while 31/4s were still doing this kind of thing, and it always seemed to me that it would have made more sense to do it the other way round.

I don't see what point you're trying to make with the Waterloo-Exeters. If a 33 needed to have the load reduced to maintain time because the ETH would have been pinching too much of the juice, then a 31/4 with less engine power available to begin with would be having the same problem only more so.

 

[Magdalia]

You sure? This lot seem to reckon they had a separate machine: http://sremg.org.uk/diesel/class33.shtml

I thought I was, but I'd not seen that link before! I knew that class 33 ETH was generated dc, but not aware that class 33 had 3 generators, including one specifically for ETH. Thanks for that, I like to learn new things.

I agree the ETH index of the 31/4s was remarkably high, but beyond their original purpose, there wasn't much call for it (in terms of booked duties, as opposed to standing in for something bigger that had gone phut).

No, the ETH index of 66 was required for the original purpose. It was required by the locos hauling the aircon sets in passenger service, therefore it was also required by the locos making the ecs movements. Don't forget that we are talking pre-HSTs here, and the Inter City service operated in a very different way: in 1974 Hornsey kicked out 10 aircon sets at start of day, that's 10 sets needing full ETH, and some of them load 14.

I don't see what point you're trying to make with the Waterloo-Exeters. If a 33 needed to have the load reduced to maintain time because the ETH would have been pinching too much of the juice, then a 31/4 with less engine power available to begin with would be having the same problem only more so.

My point is that any criticism of class 31/4s performance, when providing ETH to long trains, is also valid for class 33s, it is only a matter of degree. And the difference is not that great: 5% on horse power and about one coach worth on weight.

Once class 50s took over the Waterloo-Exeters in May 1980, class 33s deficiencies were more exposed, up trains with class 33s subbing for class 50s were normally at least 25 late into Waterloo. I know because I could see them from my office window.

 

[randyrippley]

Why didn't they just fit ETH to more 27s?

 

[Magdalia]

Why didn't they just fit ETH to more 27s?

Class 27s were not included in the Heavy General Overhaul programme and all were withdrawn by 1987. They weren't going to be around for long enough.

 

[randyrippley]

Class 27s were not included in the Heavy General Overhaul programme and all were withdrawn by 1987. They weren't going to be around for long enough.

but they could have been, in preference to the 37 and 31
Boiler out, room to fit a diesel genny for the ETH
Plenty of spare power units from the 25 fleet
You would have ended up with something close to a 33 but with the right RA

 

[43096]

but they could have been, in preference to the 37 and 31
Boiler out, room to fit a diesel genny for the ETH
Plenty of spare power units from the 25 fleet
You would have ended up with something close to a 33 but with the right RA

If they were going to do that they’d have kept the 27/2s so fitted once they were replaced by Shove-Duffs on Edinburgh-Glasgow. They didn’t; reportedly the gensets had reliability issues and we’re prone to catching fire.

 

[Magdalia]

but they could have been, in preference to the 37 and 31
Boiler out, room to fit a diesel genny for the ETH
Plenty of spare power units from the 25 fleet
You would have ended up with something close to a 33 but with the right RA

The question of what to include in the HGO programme was decided first. Only the classes that had got through the qualifiers could compete in the final!

I don't know the internal layout of a class 27 but I suspect that, for weight distribution purposes, the boiler is at the opposite end to the main and auxiliary generators. I doubt that a generator/alternator driven off the main power unit could occupy the same space as the boiler.

The ETH class 27 had a separate Deutz diesel engine to power the ETH and only produced an ETH index of 28, even lower than the class 37/4.

Power unit spares might not have been a problem for keeping class 27 going but what about the GEC traction equipment? The Sulzer Type 2 that "made it through the qualifiers" was class 26, which like class 33, had Crompton Parkinson traction equipment. The Scottish Region had experience of all of classes 24-27: they would have had significant input into the selection of class 26 for HGO.

 

[47827]

The likes of sprinterisation as well as the building of a few other DMU classes (like pacers) along with the building of other locomotive classes and a fall in freight traffic combined certainly allowed several classes BR wasn't so keen keeping on, for all sorts of reasons, to be withdrawn earlier than the usual design life of the other classes. The 27s, along with 25s, the remaining peak classes and 40s were all able to be dispensed with over just a few years. A few classes were all or largely gone by the 70s and early 80s from previous projects and arrivals too.

The 27s were OK when they worked but regarded as more troublesome than some of the others thus falling out of favour. I think the arrival of 37/4s was also a stop gap pre 156s/158s that released enough 37/0s (and a few sulzer type 2s that found their way on stuff they shouldn't still) off passenger trains to help with getting rid of smaller sulzers elsewhere.

Survival of more freight and no sprinter orders for another decade (money diverted elsewhere) and the picture may have been different with 27s and other classes surviving plus more 37/4s perhaps.

 

[Stathern Jc]

The main reason I recall for 31s needing a higher index was that after inception they were booked onto some of the long Intercity air con ECS moves into the likes of Kings Cross and needing to power the set in the sidings plus on the way to the station to ensure all was ready for the passengers before the train loco was buckled on. The easy fix to help a 31 in need of some grunt was to give it a fellow blue star pal to plug into and shift the train faster (usually another 31 but not always). Though BR did kick them out on all manner of unsuitable trains often resulting in odd incidents of killing em en route.

47827's mention of "killing em en route" brought back the memory of travelling on an "All Line Rover" in 1986. I was on a Summer Saturday Manchester to the West Country train between Birmingham and Bristol with a pair of 31s on a long rake of aircon stock (remember it was something unusual, perhaps all First Class from WCML weekday runs?). One of them had expired, and the other pulling a good load and it's dead pal managed to make a lot of noise but not much progress up the modest climb after Bristol Parkway.

As a contrast the next leg of my journey to Cardiff was with a pair of 33s with four coaches (a 4TC set?), at what seemed to be not much more than a fast tickover they went like the wind, including the climb out of the Severn Tunnel. I always enjoyed a good run with a 33.

 

[Peter Sarf]

47827's mention of "killing em en route" brought back the memory of travelling on an "All Line Rover" in 1986. I was on a Summer Saturday Manchester to the West Country train between Birmingham and Bristol with a pair of 31s on a long rake of aircon stock (remember it was something unusual, perhaps all First Class from WCML weekday runs?). One of them had expired, and the other pulling a good load and it's dead pal managed to make a lot of noise but not much progress up the modest climb after Bristol Parkway.

As a contrast the next leg of my journey to Cardiff was with a pair of 33s with four coaches (a 4TC set?), at what seemed to be not much more than a fast tickover they went like the wind, including the climb out of the Severn Tunnel. I always enjoyed a good run with a 33.

The hardest I ever heard 33s work was pairs bringing oil tanks off the Hoo Peninsular Past Hoo Junction and onwards to London. Granted it might have been the bogie gravel trains - Marinex and Brett Marine are two brandings I recall (1974 onwards).

I vaguely recall being told that the 33s had quite a good maintenance schedule to help avoid failures clogging up an EMU filled network. But it might not be true. The 33s also had good generators and traction motors.

 

[Magdalia]

I vaguely recall being told that the 33s had quite a good maintenance schedule to help avoid failures clogging up an EMU filled network. But it might not be true. The 33s also had good generators and traction motors.

I have heard a similar story, that SR had their own exam calendar, required class 33s back at home depot for exams every 3-4 days, and didn't want other depots involved. My understanding is that, even when working to Milford Haven and Manchester, locos were worked out from and back to Eastleigh every 3-4 days, via Bristol and Cardiff. It would be nice to have some details from people who know.

 

[CW2]

As regards class 27s, by the early 1980s overhauls on them had ceased, and they were in a downward spiral of poor availability, minor maintenance, and riot-a-day failures.
They had thoroughly proven themselves to be an unreliable bit of kit, which everybody was keen to see the back of as quickly as possible.
Their performance on the Edinburgh - Dundee circuit, for example, scarcely went a full day without one of the locos failing, or requiring to be swapped out of diagram.
Personally I quite enjoyed them, as the quintessential Scottish Type 2 (along with the class 26), but professionally (I was a Controller in Glasgow at the time) they were a constant liability.
There is no way that anybody with a sound mind and any experience of working with those unreliable beasties would have contemplated prolonging their lives with more ETH conversions.

Re 33s, yes I too believe they were on cyclical maintenance, whereby they returned to a nominated depot every few days. This served them well when they were confined to the Southern, but was much harder to enforce when they were escaping to North and West Wales, Manchester etc.

 

[Richard Scott]

My point is that any criticism of class 31/4s performance, when providing ETH to long trains, is also valid for class 33s, it is only a matter of degree. And the difference is not that great: 5% on horse power and about one coach worth on weight.

Maybe not a big difference on paperboy significant in the rest world. 33s had slightly lower gearing and 5 stages of field weakening. Fewer axles means less friction too. A now retired manager who's remit included Cardiff to Portsmouth said if a 33 turned up all was good, a 37/4 and it would keep time, a 31 and you threw the timetable out of the window!

 

[Magdalia]

Maybe not a big difference on paperboy significant in the rest world. 33s had slightly lower gearing and 5 stages of field weakening. Fewer axles means less friction too. A now retired manager who's remit included Cardiff to Portsmouth said if a 33 turned up all was good, a 37/4 and it would keep time, a 31 and you threw the timetable out of the window!

But I wasn't talking about Cardiff to Portsmouth. You are right that the different gearing and stages of field weakening affect performance, and that made the class 33 more suitable for a route like Bristol-Portsmouth. But the low gearing, low maximum speed, and low ETH index all make class 33 unsuitable as a Type 4 substitute on long trains with ETH, which is what I was talking about.

 

[Richard Scott]

But I wasn't talking about Cardiff to Portsmouth. You are right that the different gearing and stages of field weakening affect performance, and that made the class 33 more suitable for a route like Bristol-Portsmouth. But the low gearing, low maximum speed, and low ETH index all make class 33 unsuitable as a Type 4 substitute on long trains with ETH, which is what I was talking about.

Thought you were just comparing a 33 to a 31 and stating there was little difference whereas in reality there was. A 33 wasn't a bad substitute on Waterloo to Exeter trains (substitutingfor a type 4), I had a few, but bet a 31 would have been a total disaster.

 

[47827]

Thought you were just comparing a 33 to a 31 and stating there was little difference whereas in reality there was. A 33 wasn't a bad substitute on Waterloo to Exeter trains (substitutingfor a type 4), I had a few, but bet a 31 would have been a total disaster.

A 31, even with the ETH switched off would have needed a couple of coaches removed and would no doubt have still dropped some time due to struggling to get anywhere near the top speeds of a 47/50 the timings were geared toward. About load 4 or 5 would be the most if you wanted any heat and not too much time lost. A 37, even on a theoretical 80mph top speed would probably be better than a 31 and a 37/4 even on load 9 could just about keep everyone warm too on the route we called the Mule. The acceleration being not so bad out of stations and in those days a little of a creative fall over the 80mph top speed and you would only be losing odd minutes between stations as opposed to a 31 which would cause carnage on any single line bits. Nevertheless 37/4s to my knowledge (and most of my source is the fact I used to keenly read the workings info in mags plus know some of those who bashed the line too) didn't really work those trains. The class wasn't uncommon in the last few years but it was generally no heat, often unrefurbished, 37s that worked and often not through. Sometimes there may be swaps at Salisbury with the 37s only doing a spin to Exeter and back with something else then going into Waterloo on the return.

 

[norbitonflyer]

The ETH class 37/4 was very much a particular solution to a particular need. That was for an ETH locomotive meeting the RA5 restriction to work the Scottish branches and the Cambrian.

There was one class already capable of meeting that requirement, and with an ETH of no less than 66. Somehow, the Eastern Region weren't keen on releasing them!

Electric Train Heating -or more correctly nowadays ETS for Electric Train Supply, i.e. the provision of electricity from the loco to the coaches to power the heating / lighting / air con etc.

Heating and aircon yes, but a train hauled by a steam heat loco still had lights. They are, if I understand it, driven by dynamos on each carriage, with a battery for when the train isn't moving. (I am happy to be corrected on this)

== Doublepost prevention - post automatically merged: 19 Jan 2022 ==

Thought you were just comparing a 33 to a 31 and stating there was little difference whereas in reality there was. A 33 wasn't a bad substitute on Waterloo to Exeter trains (substitutingfor a type 4), I had a few, but bet a 31 would have been a total disaster.

33s ran the Exeter line for several years, between the withdrawal of the Warships in 1971 and the arrival of the 50s about ten years later.

 

[47827]

There was one class already capable of meeting that requirement, and with an ETH of no less than 66. Somehow, the Eastern Region weren't keen on releasing them!

Heating and aircon yes, but a train hauled by a steam heat loco still had lights. They are, if I understand it, driven by dynamos on each carriage, with a battery for when the train isn't moving. (I am happy to be corrected on this)

== Doublepost prevention - post automatically merged: 19 Jan 2022 ==


33s ran the Exeter line for several years, between the withdrawal of the Warships in 1971 and the arrival of the 50s about ten years later.

Steam heat Mk1 and early Mk2 stock has a dynamo belt system separate from the heat where energy was stored up in batteries and charged every time the train moved. An early prelude to self charging hybrids effectively. But the same system was retained for ETH fitted stock of the same type. The later IC stock and air cons have motor alternator tech which uses much more power and contributes toward the higher index of such stock.

 

[43096]

There was one class already capable of meeting that requirement, and with an ETH of no less than 66. Somehow, the Eastern Region weren't keen on releasing them!

The other regions wouldn’t have been keen on taking them given the high cost of maintaining them and questionable reliability.

 

[Magdalia]

A 33 wasn't a bad substitute on Waterloo to Exeter trains (substitutingfor a type 4), I had a few, but bet a 31 would have been a total disaster.

This comment exemplifies my point. You say that class 33 "was not a bad substitute on Waterloo-Exeter trains" even though Waterloo arrivals were almost invariably at least 25 late.

But you "bet a 31 would have been a total disaster", even though class 31s have a "not bad" record substituting for Type 4s in East Anglia and on GN commuters. I have looked through my notes and, after excluding 2 days when there were unrelated delays, I have arrival time records for 20 instances of class 31/4 substituting for Type 4s in winter on heavily loaded morning trains from Peterborough. The average late arrival is 9 minutes, and they usually concluded with dashes down the hill from Potters Bar at speeds in excess of 90mph.

 

[The Chimaera]

This comment exemplifies my point. You say that class 33 "was not a bad substitute on Waterloo-Exeter trains" even though Waterloo arrivals were almost invariably at least 25 late.

But you "bet a 31 would have been a total disaster", even though class 31s have a "not bad" record substituting for Type 4s in East Anglia and on GN commuters. I have looked through my notes and, after excluding 2 days when there were unrelated delays, I have arrival time records for 20 instances of class 31/4 substituting for Type 4s in winter on heavily loaded morning trains from Peterborough. The average late arrival is 9 minutes, and they usually concluded with dashes down the hill from Potters Bar at speeds in excess of 90mph.

Class 31's were more suited to the flat lands of the East where eventually they must have been able to get up to a reasonable speed. In area with hills and frequent speed restrictions they would have been a liability as acceleration and adhesion wasn't their strong point.

 

[Cowley]

Nevertheless 37/4s to my knowledge (and most of my source is the fact I used to keenly read the workings info in mags plus know some of those who bashed the line too) didn't really work those trains. The class wasn't uncommon in the last few years but it was generally no heat, often unrefurbished, 37s that worked and often not through. Sometimes there may be swaps at Salisbury with the 37s only doing a spin to Exeter and back with something else then going into Waterloo on the return.

I spent a fair bit of my youth bashing on the Mule and those were pretty much my observations too. I don’t ever recall a 37/4 working a passenger service on the line although the occasional 37/0 was pushed into service if there was literally nothing else around.
I guess 33s were more commonly available (there were nearly always ones kicking around at Salisbury and Exeter).

 

[CW2]

Class 31's were more suited to the flat lands of the East where eventually they must have been able to get up to a reasonable speed. In area with hills and frequent speed restrictions they would have been a liability as acceleration and adhesion wasn't their strong point.

Whenever a 31 was tied onto a decent load (say 10+) then you could expect late running and late arrivals - or boiling itself to a complete standstill. In my experience low adhesion was seldom a problem, as there wasn't enough power being transmitted to the rail to cause any spinning.

I never had a 37 on the Mule - the crews nearer London wouldn't have signed 37s in any case, so hence a 33 would always be preferred.

 

[The Chimaera]

Whenever a 31 was tied onto a decent load (say 10+) then you could expect late running and late arrivals - or boiling itself to a complete standstill. In my experience low adhesion was seldom a problem, as there wasn't enough power being transmitted to the rail to cause any spinning.

I never had a 37 on the Mule - the crews nearer London wouldn't have signed 37s in any case, so hence a 33 would always be preferred.

Adhesion was a big problem with them on hills after a shower of rain, the notched power settings meant that you couldn't finely adjust the power output to match the available adhesion and the A1A wheel arrangement made it worse because not all of their over generous weight was actually on the powered axles.

 

[Cowley]

Adhesion was a big problem with them on hills after a shower of rain, the notched power settings meant that you couldn't finely adjust the power output to match the available adhesion and the A1A wheel arrangement made it worse because not all of their over generous weight was actually on the powered axles.

I was pretty much stood next to the person who was filming this.

 

[Magdalia]

Adhesion was a big problem with them on hills after a shower of rain, the notched power settings meant that you couldn't finely adjust the power output to match the available adhesion and the A1A wheel arrangement made it worse because not all of their over generous weight was actually on the powered axles.

Class 31s were ok for adhesion most of the time but did do leaf fall adhesion problems before it was a thing with EMUs, and rain could occasionally be a problem too. In the early 1980s the start out of Settle on the morning Carlisle train during autumn was notorious. A good driver checked that the sandboxes were full and working. A few years ago D5830 visited the KWVR and attempted its first run in the rain with no sand and there was a lot of slipping. Once the sandboxes had been filled D5830 was fine.

Whenever a 31 was tied onto a decent load (say 10+) then you could expect late running and late arrivals - or boiling itself to a complete standstill.

Boiling over was an issue in hot weather, but usually when number 1 end was leading. Locos rarely boiled over when number 2 end was leading. Most instances were triggered by shutting off power for station stops after long periods of full power running at high speed. The cooling system was designed for a 1365hp Mirrlees.

 

[CW2]

I think we can conclude that if the answer is "Yes, a class 31/4 is ideal for that" then you've probably asked the wrong question.

 

[Magdalia]

I think we can conclude that if the answer is "Yes, a class 31/4 is ideal for that" then you've probably asked the wrong question.

In the late 1970s and early 1980s the answer "yes xxx is ideal for that" applied in not very many cases. It was all compromises to make best use of what was available on a limited investment budget. Which takes us back to your original question....