Today is...
Friday, May 25, 2018
Welcome to, the global online
community of automation professionals.
Featured Video...
Featured Video
Watch an animation of a conveyor stacking operation demonstrating the use of a move on a gear command.
Our Advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive
Detroit diesel engine test procedure
please help me in developing test procedures for detroit make diesel engines used to drive Frame-5 GE gas turbine.

Dear All

We are using detroit diesel engines to drive our gas turbines during startup.

it has been found that many diesel engines are getting seized. so we are sending them to local vendor for rebuilding.

I need detailed test procedures to make sure the rebuilt engines are in acceptable condition.

Thanks in advance

It all depends on what you, as the purchaser of the rebuild, define as acceptable. Is it that the engines run at rated speed and produce rated torque? That they also run at the necessary speeds defined by the fuel controls used on the turbine?

How does one test an engine? It produces torque, so if an engine is rebuilt the best method of ensuring it was rebuilt properly is to put it on a dynamometer and see if it produces the nameplate rated power output.

The other test would be a compression test to see if the compression for each cylinder is at or near rated for the engine and configuration after it was rebuilt.

If you then want to test it so see if the linkages used to control the fuel rack and the shutdown are working properly, you need to define what the speeds should be for each of the solenoid states, run the unit, and make any adjustments as necessary. This would likely have to be done after the dynamometer or compression tests, and after the engine was reinstalled on the Accessory Base, to use the Speedtronic outputs to energize and de-energize the solenoids. When the engine is installed on the Accessory Base, adjustments could be made as necessary to the fuel rack or control elements.

Just remember: The cost of any testing will be included in the price of the rebuild. And, you can't ask for testing to be done after the refurbishment has been completed without a requirement in the original contract for testing, and an agreement on what constitutes acceptance or failure (which happens a lot in a certain region of Planet Earth).

Before you keep seizing and rebuilding engines, have you found the cause of the seizure of your engines?

A couple of points to check:

1. Is the lube oil level in the diesel engine sump at the working level before you start the engines. These diesel engines like all others consume lube oil, and unless replenished, the engine will end up running without lube oil.

2. Are the engines being filled up with the right grade/quality of lube oil. Grade is dependent amongst other things on site ambient conditions, mainly temperature.

3. Is the diesel engines' low lube oil pressure trip functional?

4. Is the diesel engines' cooling OK, is there appropriate flow of cooling water? You can seize a diesel engine if it overheats due to insufficient cooling.

5. Is the diesel engine speed governor properly set. Is the maximum speed/power setpoint set as per manufacturer's specification?

It is not normal that these diesel engines seize, so there must be a cause for your problem.

Just would like further explain the reason why i am asking for a test procedure.

The engine rebuilder is claiming that his engine is not the cause for this problem, so we want to have a test procedure after rebuilding the engine and installing it in the turbine base which will be an agreement for accepting the engine.

For example, would you accept the engine condition if the diesel engine disengage at the setting speed of 60% turbine speed which is 3000 rpm?

Back to the problem of engine seizing, the rebuilder suggest that the reason for this is condensation of flue gases that collects in exhaust piping then entering into cylinder liners. please advise if this is possible and how we can check if the source of problem is not from the engine side?

I thought this was the case; not original equipment failures, but failures of rebuilds. That wasn't clear from the original post, and it's still not really clear, either.

Are you experiencing a high failure rate of engines which were supplied with the turbines, or which have been rebuilt by other companies?

If the exhaust configuration hasn't changed from the original, and the original engines lasted a good many years, then why are these rebuilt engines failing because of flue gas condensation if the original engines didn't?

There's a lot of things which just aren't adding up here. It might even be the installation of the rebuilt engines that's at least part of the problem.

In university I used to work for a local, third-party "Cat-skinner" that rebuilt diesel engines of many makes and sizes. The company guaranteed their rebuilds for so many hours (I think it was 10,000 hours for most applications which were earth-moving equipment, buses and trucks, rubber-tired gantry cranes, and gensets).

We used OEM rebuild kits and brand-name gaskets, and had some really good supervisors with many years of experience who oversaw every rebuild as if it was their own engine. The only testing we did was compression testing of each cylinder and testing of fuel injectors and fuel pumps. The only engines I saw come back were two genset engines; one "exploded" on overspeed, and the other was destroyed by reverse power when the generator breaker failed to open.

Anybody can say they can rebuild engines. Did you speak to previous customers of the firm to see what their experience was?

Is the rebuilder recommending a new exhaust configuration? Was the rebuilder aware of the existing exhaust configuration and the potential for damage?

Disengagement of the starting means is a function of several things, not just diesel engine condition. Torque converter condition, acceleration rate, and diesel fuel rack adjustment can all affect the disengagement of the clutch and starting means. I believe most sequencing "drops out" the starting means at 60% speed if the clutch hasn't already disengaged by then. Again, there are several factors, not just one, that affects disengagement.

When a maintenance outage is performed on the gas turbine (which is another "engine") what kind of performance guarantee do you get? How do you verify the work was performed properly?

Again, these are diesel engines; they are basically commodity items. They are the same engines used in lots of other applications around the world. It doesn't seem to make much sense, based on what we've been told, why a testing procedure is needed. Compression is just about as good an indicator of performance as any other. It's what commonly used to assess engine performance, of just about any reciprocating engine.

If the problem is longevity, or lack thereof, testing isn't generally going to indicate that. In fact, running an engine, or every engine being rebuilt, for tens of hours under load for testing can be considered to be counterproductive, reducing the life expectancy of the engine. Again, it's going to be expensive and included in the cost of the rebuild.

We just don't have enough information about the entire situation, but I agree with bob peterson and jojo: Better to do some root cause analysis and stop the failures than to try to use testing to prevent early failures.

Maybe it's time to find another vendor. I would just like to know the whole story.

By curt wuollet on 18 January, 2010 - 8:15 pm

I would venture a guess that your engines are suffering from a lack of use. It takes several running hours after a rebuild to seat the rings and wear in the new bearings, etc. After that the engine performance should be stabilized. But you can't simply let them sit, especially in any sort of a damp environment. They need to be run periodically to keep the cylinder walls and other internal surfaces oiled, etc. This is a normal maintenance item for standby engines that are expected to be available at any time. I would think the best test, or at least a good one, would be to simply run the engine at load for a couple hours at acceptance and from start to operating temp, at least monthly. Regular running will take care of any moisture problems and will not measurably reduce the service life as industrial diesels are long life engines. You will also greatly reduce fuel system problems. About the only ways to kill them are to run them without coolant or, let them sit.

But, what do I know?


You know a lot more than you're giving yourself credit for, Curt!

Curt makes an excellent point about seating the rings and wearing in the bearings, and about the idle times involved.

These diesels are used to start the gas turbines rotating and for assistance with acceleration after flame is initially established until the turbines are "self-sustaining". The diesels usually run for about 20-30 minutes during a normal start cycle (depending on the purge timer, the diesel warm-up timer, and the diesel cooldown timer.) And, then they sit idle until the next start attempt.

For some heavy duty gas turbines, the next start could be months away. Or, it could be later the same day. Or daily. Or not for a week, or a month. Or even six months. It depends on how the turbine is being operated.

The turbine control system should have a 'Diesel Test' function, that allows the diesel to be started and operated at idle for as long as the user desires. And this should be a regular, periodic maintenance activity for gas turbine starting diesels, just as it is for emergency power diesel gensets.

And, a proper run-in is also necessary for seating new rings in new or honed cylinders, and would help with "running-in" the bearings as well. So, as part of the rebuild it might be a part of the spec to run the diesels at rated speed (and load) for some period of time to seat the rings, based on a recommendation from the rebuilder.

If the rings are not being allowed to seat properly, this could be the cause for at least some of the seizures, if not all. Perhaps after the diesels are installed, they should be used for cranking the turbines for a couple of hours, at a minimum, to allow the rings to seat and the bearings to run-in. It's going to be costly to get a load connected (and disconnected) to the diesels in the rebuilder's facility.

Bob Peterson and jojo and cww are correct: Fix the root cause(s) of the failures and you won't have them. A test procedure isn't going to identify and fix the root causes, and sometimes that can take a lot of analysis and data-gathering to do. But, in the long run it's the best choice.

Please write back and let us know how you proceed. "Feedback is our most important contribution"(c) here at as it lets others know who read this post next week, or next year, if the information provided was useful or not.

By bob peterson on 18 January, 2010 - 10:08 am

Many? Diesel engines don't typically just seize up. If you are having this problem, it is probably more important to figure out why it is happening and fix it than to worry all that much about what test procedure is used after rebuilding.

As for a test procedure, I think I would ask the manufacturer of the engine. They probably can give you some ideas.

At the very least I would want the engine put on a dyno and run across what its speed range is supposed to be at varying loads. Probably in your case it only needs to run at whatever its idle speed is (zero load), and its normal speed (zero to 100% load, or maybe 100+% of full laod depedning on how your engine is rated).

I would want to see a running graph of the speed, the load, and the fuel usage for all conditions tested, along with air intake temperature and
exhaust temperatures at each cylinder.

I would also want NEW oil put in (don't laugh about some rebuilder not replacing the oil, or reusing old oil) and samples taken at various points in the test cycle and sent to a reputable lab for analysis. Maybe at 0, 1 ,5, 10 and 20 hours into the test.

You should probably also insist they tear the thing down at some point after it has run for 10 or 20 hours after a rebuild and inspect the bearings and shaft for any wear.

Some further questions:

1. How often do you run your diesel engines (i.e. start-up your gas turbines)?

2. How long are the gas turbines kept idling (i.e. not used) during the year?

3. Do the diesel engines seize up while they are running up the gas turbines, i.e. somewhere in the middle of the starting cycle, or when you try to crank the diesel engine (i.e. at the beginning of the starting cycle), you find the diesel engine seized up?

4. Are the gas turbines installed in a harsh environment?

5. Is the fuel being fed to the diesel engines of the right quality?

Small comment, the break away speed stands around the 60% mark you are indicating. This speed does not depend on the diesel engine, but on the point where the gas turbine is trying to overspeed the diesel engine (i.e. it can generate enough torque to try to push power into the diesel engine).

Please find the answers for your questions

> 1. How often do you run your diesel engines (i.e. start-up your gas turbines)? <

diesel engines are kept idle till its required to start the gas turbines so its difficult to know exactly the operating frequency of the engine.

> 2. How long are the gas turbines kept idling (i.e. not used) during the year? <

Because these gas turbines are standby and used only when there is shortage of power so these units could be kept idle for more than a month

> 3. Do the diesel engines seize up while they are running up the gas turbines, i.e. somewhere in the middle of the starting cycle, or when you try to crank the diesel engine (i.e. at the beginning of the starting cycle), you find the diesel engine seized up?

Before these engines will sieze up we have observed a lot of water ( or it could be condensed gases) from the engine exhaust which is connected to the gas turbine exhaust duct.

> 4. Are the gas turbines installed in a harsh environment? <


> 5. Is the fuel being fed to the diesel engines of the right quality? <

I guess you are suspecting the diesel fuel could have condensed water in it which is possible.

Would like to share with you this point. Drain tubes is connected to the diesel engine exhaust piping and was told that it was fixed by the company which installed the units. I really don't know what is the use of drain tube if the engine exhaust should be free from condensates?

Alot of good advice given so far. the original poster said something about condensate..and cylinder liners.

this makes me wonder if the exhaust from the diesel is routed so that it is common with the turbine exhaust at some point...

this would tend to cause liquids to run down into the exhaust manifold and possibly into the cylinders...
maybe I just did your root cause analysis(?)

let us know.

> I really don't know what is the use of drain tube if the engine exhaust should be free from condensates? <

One of the exhaust products of an internal combustion engine is water vapor. When an internal combustion engine is started from a cold condition, it's quite common to see water draining from the exhaust. This is because the exhaust components are cold and cause the water vapor exiting the cylinder in the exhaust gases to condense. Once the engine has been run sufficiently to warm the exhaust system, the water vapor will no longer condense and will exit the system as "steam" (water vapor).

So, the purpose of the drain line is to allow a path for any condensed water vapor to exit the system without draining back to the engine, and possibly being drawn or drained back into the cylinders (through the exhaust valves) where it can form rust on the cylinder walls.

[That's why it's best to not make short trips with cars and trucks, and to allow the "engine" (including the exhaust system) sufficient time to warm and evaporate any condensed water vapors so they don't rust the exhaust system.]

The exhaust from most starting means diesel engines goes up through the roof, into a muffler, and then out to atmosphere. Usually, the muffler is a larger diameter than the exhaust pipes entering and exiting the muffler, and the drain line is connected to the muffler, since that's the low point of the system. It should be free and clear to allow condensed water vapors from any source to drain. Some diesel exhaust were improperly constructed and rain water was found to enter the system, but also draining out of the drain line.

In humid environments, humidity can be "drawn" into a cooling exhaust system after an engine has been run and they can also condense. But the volume shouldn't be very much at all.

You really didn't answer the question about when the seizures occur. Are they occurring when the diesel is running or has been running for a few minutes, such as during firing and acceleration of the unit after flame has been established? Or are the seizures occurring shortly after starting, such as during the starting means diesel warm-up period when the RPM is low, or during acceleration of the turbine shaft from zero speed during cranking when load on the diesel is increasing?

You say a "lot" of water has been seen coming from the exhaust before the seizures, but is that water observed coming from the starting diesel exhaust drain line, or where is it being seen to come from?

I've never seen a starting diesel exhaust connected to a gas turbine exhaust duct, either. Because that would "back-pressure" the diesel exhaust system by the gas turbine exhaust when the gas turbine was running. This could cause moisture in the gas turbine exhaust to condense in the starting diesel exhaust system when it was back-flowing through the drain line. the flow wouldn't be high, but over time, it could result in considerable condensation. I've seen starting diesel exhaust pipes (from the muffler exit) run up the side of the gas turbine exhaust stack, but never into the gas turbine exhaust stream, or into the exhaust duct at some point below the top of the stack. (I've seen L.O. Tank vents run into the exhaust stream, but not diesel exhausts.)

Again, it's not uncommon to see condensed water vapor coming from the exhaust of an internal combustion engine for a short period of time after it is started from a cold condition. Until the exhaust warms up, the water vapor in the exhaust gases will condense, and they need a path to drain *away* from the engine. So, some condensate is practically unavoidable.

We don't have enough information:

1) about the volume of condensate seen coming from the exhaust prior to the seizures or where it is being observed and when it is being observed;

nor do we fully understand:

2) when the seizures are occurring.