[andrewkloppel]

I have noticed a couple different times now that after going away for a long weeked, i.e. anything 3 days or more, I come back to find that the car is very hard starting.

Starter is still turning rapidly, no battery issues, but I will have to sit there for a good 2-3 minutes of holding the key over for 10 seconds at a time until the thing will finally turn over, which thankfully it still will start after some trying.

ASide from this trouble after it sits for some time, the car is a starting champ, always turns right over, and once I get this first hard start out of the way after some sitting, the car starts up every day to and from work like a charm and continues to start easily until that long weekend comes along every once in a while, then I have the same problem.

Fuel sitting in the lines somehow gumming up? I did install a new fuel filter about 1K ago, and like I said, aside from when it sits for a 3 days or more, starts like a charm. Any thoughts on possible issues?

 

[sbaert]

1st step is replacement of the OVP relay and ALL the fuses with new genuine MB replacements.

On your model year, the OVP is tied into the fuel pumps.

 

[andrewkloppel]

Any reason you could think that it would be an intermitent issue?

My understanding is that the OVP is an electrical component, making me think that (like a fuse) when it goes out, its out, not that it might slowly fade after a few days sitting.

Can an OVP deteriorate over time causing something liek my symptoms without completely being bad (i.e. it works fine the rest of the time)?

 

[sbaert]

The OVP relay is a wear 'n tear item and is crucial to having a proper functional electrical system. I replace my OVP whenever I put in a new battery.

The fuses corrode internally and develop hairline cracks invisible to the human eye.

Replace both and go from there.

BTW, what was the outcome on your alignment problem???

 

[andrewkloppel]

Ok thanks, I'll look into obtaining a new OVP. Im guessing you're going to tell me the Uro brand is garbage and only to buy OEM?

Alignment issue was never resolved. I recently got engaged and have been dedicated to that. Not to mention the fact that moving into winter here in Washington state means I now leave for work in the morning in the dark and return home in the dark and have no garage to work in when I get there.

The last alignment Sears did for me got everything "in the green" and at least levelled the steering wheel. I still have a moderate pull to r/h side at most speeds, but after paying a lot of attention to my steering for that period of time, realized that my steering is actually feeling somwhat loose at freeway speeds and sometime will drift either left or right...

Is there any remedy for "loose" steering? Like a magic bolt I could tighten and then everything would be ok? Because that's about all I would have time to work on these days it seems...

 

[sbaert]

URO = Unreliable Reproduction Offering. Get the KAE(hler) relay from autohausaz.com or genuine MB. The KAE relay is only about $50 IIRC.

As far as your steering, if your steering components check out OK I would seriously look into checking for play at the steering shaft coupler joint as well as play in the steering box.

Be very careful with the steering box preload adjuster nut as it can result in your steering locking up without warning.

 

[andrewkloppel]

Where are the shaft coupler and box located at? Are these things I could check with a visual inspection or is some teardown needed?

 

[sbaert]

They're located below the brake booster, just above the frame rail.

Only accessible from below.

Just follow the factory service manual to determine if the steering wheel play is still within spec.

 

[jftu105]

1st step is replacement of the OVP relay and ALL the fuses with new genuine MB replacements.

On your model year, the OVP is tied into the fuel pumps.

I have the circuit diagram for E320, 1995 in front of me. OVP does not have any direct connection to the fuel pumps. Both OVP and fuel pump relay are connected to battery positive terminal, along with many other devices. The fuel pump relay is controlled by ECU. The fuel pumps are controlled by the pump relay.


jftu105

 

[bmwpowere36m3]

I have the circuit diagram for E320, 1995 in front of me. OVP does not have any direct connection to the fuel pumps. Both OVP and fuel pump relay are connected to battery positive terminal, along with many other devices. The fuel pump relay is controlled by ECU. The fuel pumps are controlled by the pump relay.


jftu105

True, on the '95 E320 there is no direct connection between the OVP and the fuel pump relay. The signal for the FP relay comes from the ECU, but the OVP does trigger something within the ECU. So whether they are indirectly related I don't know (would need a schematic of the ECU).

Either way, can't hurt to pull the OVP and run the basic checks on it with a DVM.

Pin 1: +12 V hot
Pin 3: +12 V switched (ignition)
Pin 5: Ground

With Pin 1 and Pin 5 connected to +12 V and ground, respectively. Only Pin 9 should have +12 V (any other pins have +12 V = OVP no good). Pins 4 & 6 should be open (infinite); one DVM lead on Pin 4, other on Pin 6 and DVM on ohm scale (any reading, ohms = OVP no good).

Apply +12 V to Pin 3 (relay should click), Pins 9/8/7/2 +12 V (any pins not reading +12 V = OVP no good). Pins 4 & 6 should be closed (shorted); one DVM lead on Pin 4, other on Pin 6 and DVM on ohm scale (reading great than a few ohms = OVP no good).

 

[Floobydust]

Andrew -

My personal belief is that your starting issue is not related to either the OVP or the fuses, although it is a good idea to prophylactically clean/replace the original tin/lead element fuses. These were an engineering abomination from a reliability perspective.

Intermittent electrical failures are rarely as reproducible as you have described, so I would suspect something more mechanical like the fuel pumps themselves, the pump check valve, or the fuel pressure regulator. Fuel pressure leak-down or pump priming failure can be a very consistent scenario.

- FD

 

[Floobydust]

With DVM on diode mode, Pin 1 (DVM negative lead, black) and Pin 5 (DVM positive lead, red) it should read around 0.4-1.5 V (a reading outside this range = OVP no good). Reverse the leads and it should read infinite or open (any voltage reading = OVP no good).

A 1995 E320 will be equipped with a 000 540 67 45 protection relay and a single 15 amp fuse. The "diode test" you describe above will always fail on this relay because it contains no zener diode like the OVP relays that came before it. The diode test is applicable only to the 5-pin, 7 pin, and the dual fuse 9-pin relays.

- FD

 

[bmwpowere36m3]

A 1995 E320 will be equipped with a 000 540 67 45 protection relay and a single 15 amp fuse. The "diode test" you describe above will always fail on this relay because it contains no zener diode like the OVP relays that came before it. The diode test is applicable only to the 5-pin, 7 pin, and the dual fuse 9-pin relays.

- FD

I might be wrong as far as the diode test, but I have that exact OVP and doing the diode check I got a reading of 0.7 V and reversed nothing… Looking at the schematic on my OVP there is no symbol for a zener, just a regular diode. So if it doesn't have a zener, then there is no "over voltage" protection in this OVP, right?

 

[Floobydust]

I might be wrong as far as the diode test, but I have that exact OVP and doing the diode check I got a reading of 0.7 V and reversed nothing… Looking at the schematic on my OVP there is no symbol for a zener, just a regular diode. So if it doesn't have a zener, then there is no "over voltage" protection in this OVP, right?

I have dismantled and derived the complete internal schematic for this relay (and the others) using a genuine, original equipment Mercedes Benz relay. There is indeed no internal zener diode nor is there any other provision for over voltage protection in this relay. So yes, it does not provide a true over voltage function, but it does provide protection via the 15 amp fuse which blows ins response to the current shunting of the internal over voltage protections in the down stream protected devices. In that sense, this version of the OVP is really just a fuse protected, combination relay.

The "diode" that you are measuring is the inductive kickback protection diode for the internal transistor relay coil driver. The current flow during this measurement is through the K2 relay coil, so the measured values could be affected by the resistance of the coil depending on the type of meter you have.

- FD

 

[sbaert]

If the vehicle is hard to start after sitting more than 12 hours, then the most likely culprit is a bad fuel regulator, which is somewhat common on the M104.

 

[bmwpowere36m3]

I have dismantled and derived the complete internal schematic for this relay (and the others) using a genuine, original equipment Mercedes Benz relay. There is indeed no internal zener diode nor is there any other provision for over voltage protection in this relay. So yes, it does not provide a true over voltage function, but it does provide protection via the 15 amp fuse which blows ins response to the current shunting of the internal over voltage protections in the down stream protected devices. In that sense, this version of the OVP is really just a fuse protected, combination relay.

The "diode" that you are measuring is the inductive kickback protection diode for the internal transistor relay coil driver. The current flow during this measurement is through the K2 relay coil, so the measured values could be affected by the resistance of the coil depending on the type of meter you have.

- FD

Awesome info, thanks. I wonder if due to all the issues with OVPs over the years, MB gave up on providing over-voltage protection. Like you said, nothing more than a fuse protected relay.

 

[jftu105]

If the vehicle is hard to start after sitting more than 12 hours, then the most likely culprit is a bad fuel regulator, which is somewhat common on the M104.

Didn't you suggest replacing OVP as the first step?

 

[Floobydust]

Awesome info, thanks. I wonder if due to all the issues with OVPs over the years, MB gave up on providing over-voltage protection. Like you said, nothing more than a fuse protected relay.

The primary reason for the change was that automotive electrical systems were becoming more complicated with time. MB realized that it would be impractical to continue the "central point of protection" concept going forward. So starting with the more sophisticated sequential fuel injection systems, MB took a "distributed" protection approach for surge and over voltage protection. A primary switching relay was still employed, but the modules had to fend for themselves. The V8 cars used something MB refers to as a "base module" whereas later cars went to something MB refers to as the K40 relay.

MB also improved the reliability on the base module and the K40 by eliminating one of the two primary causes of failure for the OVP. MB unwisely chose to locate the original OVPs in the open engine environment. The cheesy outer plastic cover really did not offer the kind of environmental protection the OVP really needed. MB remedied this situation in the base module and K40 cars by enclosing them in a weather tight enclosure, much like the relay box at the back of our fuse box. This significantly improved the reliability of these units even though the internal construction of the base module and K40 relay is fundamentally the same as the OVP relays (minus the zener function). They all use the same internal type relay and they are all of PCB soldered construction. The solder connections, by the way, are the second of the two weak points, a problem which which is common to almost any module of sufficient age.

- FD

 

[jftu105]

True, on the '95 E320 there is no direct connection between the OVP and the fuel pump relay. The signal for the FP relay comes from the ECU, but the OVP does trigger something within the ECU. So whether they are indirectly related I don't know (would need a schematic of the ECU).

Either way, can't hurt to pull the OVP and run the basic checks on it with a DVM.

Pin 1: +12 V hot
Pin 3: +12 V switched (ignition)
Pin 5: Ground

With Pin 1 and Pin 5 connected to +12 V and ground, respectively. Only Pin 9 should have +12 V (any other pins have +12 V = OVP no good). Pins 4 & 6 should be open (infinite); one DVM lead on Pin 4, other on Pin 6 and DVM on ohm scale (any reading, ohms = OVP no good).

Apply +12 V to Pin 3 (relay should click), Pins 9/8/7/2 +12 V (any pins not reading +12 V = OVP no good). Pins 4 & 6 should be closed (shorted); one DVM lead on Pin 4, other on Pin 6 and DVM on ohm scale (reading great than a few ohms = OVP no good).

True, it never hurts to check the health of the OVP, if you know how to check it correctly. Your procedures are very detailed and are right for a certain types of OVPs according to Flooby.

ECU controls so many things and ECU has a direct connection to the battery positive. ECU can tell the fuel pump to pump when you turn the key before cranking without any information from OVP. OVP will have no effect before the engine starts because the voltage will be only +12V, give or take a fraction of a volt. Therefore, logic thinking will not suspect the OVP as the first suspect for this problem. It must be fuel related.

I just wanted to point out "subtly" that sound reasoning should be the first step to fix the problem, not replacing OVP as the first step for all problems. The way it goes, if I post that my car get a flat tire after sitting for a few days, I might get a suggestion to replace the OVP first.

Check out my post on "The Bad rep of OVP, or is it?" With this case included, it is 3 for 17 for the success of blindly replacing OVP.

jftu105

 

[jftu105]

The primary reason for the change was that automotive electrical systems were becoming more complicated with time. MB realized that it would be impractical to continue the "central point of protection" concept going forward. So starting with the more sophisticated sequential fuel injection systems, MB took a "distributed" protection approach for surge and over voltage protection. A primary switching relay was still employed, but the modules had to fend for themselves. The V8 cars used something MB refers to as a "base module" whereas later cars went to something MB refers to as the K40 relay.

MB also improved the reliability on the base module and the K40 by eliminating one of the two primary causes of failure for the OVP. MB unwisely chose to locate the original OVPs in the open engine environment. The cheesy outer plastic cover really did not offer the kind of environmental protection the OVP really needed. MB remedied this situation in the base module and K40 cars by enclosing them in a weather tight enclosure, much like the relay box at the back of our fuse box. This significantly improved the reliability of these units even though the internal construction of the base module and K40 relay is fundamentally the same as the OVP relays (minus the zener function). They all use the same internal type relay and they are all of PCB soldered construction. The solder connections, by the way, are the second of the two weak points, a problem which which is common to almost any module of sufficient age.

- FD

FD,

dito to your awesome information.

It answers some of my suspicion of OVP's role in the past. I wrote on the forum before why MB could not just beef up the voltage protection inside ECU and the ABS control unit but rely on OVP alone. Electronically, it is so easy to add a voltage regulation circuit inside a module (a zener diode alone offers some protection). There are voltage regulation chip available to reduce any voltage from 18V to 12V easily. ECU sees the battery positive voltage directly anyway. There is also an alternator voltage regulator to limit the voltage to 14.5V, mostly under 14V.

Let me repeat this. OVP is important but it is not the source of all the problems. Sound reasoning is more important than blind replacement of OVP.

jftu105