I never had this problem so I can't help you but try doing a forum search for "P0410" and you'll find lots of threads about it.

http://www.benzworld.org/forums/search.php?searchid=21244577

Good luck!
Fred

 

Thanks fosmith.
I did do a good search, but found that there are many possibilities including bad o2 sensors.
I was merely trying to learn from others recent experience with this CEL code.
I will perform more diagnostics and see where they lead.

 

You have an intermittent problem which makes the diagnosis to a particular part even more difficult. Sometimes other problems (like O2 sensors, EGR, improper coolant temperature, vacuum leaks etc.) can trigger this code, as the way it is detected is during the drive cycle the secondary air pump is turned on , air is injected into the exhaust side and the O2 sensor(s) are monitored for a very lean condition (due to injected oxygen in the air) in two consecutive drive cycles. Thus, you need to make sure there are no other conditions that would interfere with the test results (increase of lambda from the O2 sensors by 23 percent or more)

Once the problem is isolated to the secondary air injection system, then the operation of the air pump, switchover valve and the combination valves (one per bank) can be checked. The common problems are sticking switchover valve, leaky check valves (of the combi valve), air and vacuum supply hoses between components (kinks, holes , splits, loose connections), intermittent electrical contacts, and the intermittent operation of the air pump (which requires something like 35 amps, and possible to have worn brushes), and the air pump relay itself (worn contacts over time). The air pump turns on when the engine is cold for few minutes, so you can "simulate" the cold engine condition to check the air pump and the equipment down the chain.

 

When you start a cold engine, the secondary air pump, pumps air into the exhaust system to raise the temperature of the catalyst and thereby reducing emissions. As has been mentioned the solution can be hard to determine.

One thing you can check is whether the pump comes during a cold start and goes off after a few minutes. Also, there are three little check valves associated with the pump tubing. One or more of those could be plugged.

 

I can see a connector at the top of the air pump. Once I remove the connector I can see two terminals. One of these is connected to the body of the pump through the connector (?).
Should I be able to read a resistance between the two terminals at the top of the pump to see if the motor is still OK ?
Is there anything else I can check with the meter or other terminals I should look at ?

The vacuum hoses all look fine. On my car, it looks like there are two check valves ( hockey puck like structure ). I removed the two Philips screws on each and removed the tops. These valves had a lot of deposits on them, but the valves were not stuck.

At this stage, I am suspecting the air pump it self but I am not 100% sure how I might check that.

 

You should be able to check the resistance, between the terminals, no problem. From the electrical point of view, you have the battery supply in K40, relay module at the passenger side, 40 amp maxifuse (the orange big one) the air pump relay contacts and the air pump. It is just that you describe the problem as intermittent, so the continuity (low resistance) reading may or may not be there if the pump brushes are worn out. Just over the pump on the left side there is the switchover valve which is just a solenoid valve, that you may want to check the connector and operation. You should be able to hear the air pump running with engine cold for two minutes during start up.

 

Please look at the attached documents. This is for W163 but the same engine / air injection system. One of the problem areas is the pump relay in K40 unit, the contacts go pitted over time and eventually stop supplying the power to the pump altogether.

 

Thanks very much for this information.

I have checked the relay and the associated 40A fuse. The relay seems to work fine when the coils are activated with a 9V battery and the resistance measured at the contacts. I can do this repeatedly without issues. The check valve is also not blocked and seems to work OK.

However the pump itself exhibits a strange behavior. When I measured the resistance at the two terminals it was open circuit, so I took off the cover on the RHS of the pump that is held there by 3 Torx screws.When I remove that cover completely or replace it "almost completely", I read a very low resistance. I think that is the resistance of the pump motor. However, the terminals become open circuit if the cover is replaced completely as it it was in the first place. It may be that the brushes are worn, but I think the remedy would be a new pump ?

I guess I would have to take the pump out to be sure as noise from the engine is such that its hard to tell if the pump goes on or not.

The documents are very helpful and educational.

 

I guess I would have to take the pump out to be sure as noise from the engine is such that its hard to tell if the pump goes on or not.

The air pump makes A LOT of noise when it comes on. Even from inside the car with the hood and windows closed, its sound--a rather high-pitched hissing, a bit like a vacuum cleaner--is clearly audible. Even if you can't distinguish it at first, you will clearly hear the difference when the pump switches off after 30-60 seconds or so, given normal operation.

If you are standing in front of the open hood, having just started the cold engine, you can hear the air pump even better. You can also feel it by putting your hand on the housing. Or you can pull off the big rubber hose at the back and you will feel a strong airflow coming out of the pump where you took the hose off.

Thus, you do not have to take the pump out in order to establish if it is coming on. If you are unable to hear, observe, and feel any of this, it is all but certain that the air pump does not run. In that case, simply apply 12V to it (from a strong enough power source) and see if it actually works.

 

You can put your hand on the pump immediately after a cold start and you should feel the vibration of the working pump. Or you could check the air pump air output for 90 seconds or so the pump runs.

If you want the check the air pump operation with a warm engine, to turn the air pump on, you could use a little "trick" by removing the engine coolant sensor connector, and terminate the connector at the ECM end with a 2000 ohm resistor. The ECM will think the coolant is at 30 degrees C, so it will start the pump even though the engine is warm. This is just for testing purposes.

When you remove the check valves and had a look inside the holes at the engine side have you noticed any carbon / gunk buildup ? With both check valves removed when you start the engine you should have some exhaust gasses coming out from those openings to confirm that there are no blockages.

If you have a scanner with live data display (like graphs) you should be able to display the sharp change in the outputs of the O2 sensors when the secondary air injection pump runs. This is what the ECM observes to test the proper operation of the system.

It is hard to determine the condition of the air pump unless you open it up and check the condition of the components. Another failure mode of the pump is due to exhaust gasses entering in it due to defective check valves. When you open it you should be able to see the gunk inside.

 

I would like to re-confirm that my pump has failed.

If I measure the resistance at the two pump terminals after removing the connector, what value should I read ?
I usually read open circuit, but if I fiddle with the cover at the end of the pump by removing the three screws, I can read almost a short ( 0.6 Ohms or so ). Something in the pump is intermittent, but I am not sure if its beyond repair.

 

I suggest you activate the air pump and observe its air output. You can measure the resistance but with the low voltage and current of the ohmmeter, it is likely to be a not reliable measurement at static conditions (due to some electronics inside for start up.). The pump takes around 30 to 35 amps when it runs, so the running impedance is less than 1/2 ohm.

The pump runs few minutes per day (only when the engine is cold), so if the pump pumps air when it runs, I would look at other areas like the switchover valve, combi valves and clogged air passages from the combi valves to the exhaust.

 

I have an intermittent contact on the pump. The positive terminal on the pump's does not make contact with the strip of metal inside the cover that has the three small torx screws. I soldered the terminal and I can hear the pump coming on and going off after the first 30-40 seconds or so. However when I restart the engine after idling about 3-5 minutes, I hear the pump come on and stay on for a lot longer than 30 seconds. I think its running on open loop.

Is the time for the cold start controlled by the ECM and after the first time it has come ON and then turned off, is it controlled by another sensor that may have possibly gone bad ?

I am not sure if I found the correct relay for the pump. I can see a 40A fuse and a 4 terminal relay under a cover held down by 4 Philips screws that has ECM connections inside and this single relay and 40 A fuse. I have checked this relay with a battery and it works fine, but is this the correct relay to check ?

 

The relays are lined up like k1 k3 k2 and the relay for the air injection pump is k3,

The ECM will activate the relay during start up for around 90 seconds to 2 minutes if the coolant temperature is below certain levels (40 degrees C I think). It will also test the the operation of the system under certain condition, once per drive cycle and it turns it on few seconds or so. This is what generates P0410, not the the start up operation. There is no sensor that I know of directly controls the operation. however the ECM uses the coolant sensor data and other sensor data (like the rpms, throttle etc) to determine when to do the test. If this error condition occurs in two consecutive drive cycles, you get the CEL.

 

Perfect explanation !.
I have verified that the pump goes on at the start and goes off again after a minute or so,
My engine is fairly noisy and I thought the pump was turning on continuously, but when I looked for the output it was not running.
Now, I will watch out for the code to see if P0410 returns.
These pumps are not so expensive on Ebay, unless one wants a genuine new pump.