Gemeinsame Masse

Hi,

To the best of my knowledge, there is no standard or regulation that requires a notebook power adapter to connect the vehicle's ground to the computer's ground potential.

If that's not the case, there will be problems, specifically with the computer's motherboard.

The fact that it works on many systems is actually just an indication that most computers seem to be built in a similar way, which, to be honest, seems a bit too vague to me.

By the way, I'm not referring to 12V to 230V adapters with a 230V power supply connected to them, because those typically provide galvanic isolation. I'm talking about adapters that directly use the car's internal power supply without complex circuitry, and which either pass the ground connection through or don't, and thus have a different potential relative to the car.

Best regards,

Jan.

By the way, I mean

You're worrying about a very rare problem. Ross-Tech hasn't sold interfaces with galvanic isolation for 2 years. I can count on one hand how many people have had a ground loop problem, and in every case, it was due to an extremely cheap 12V to laptop power adapter. A decent 12V to laptop power adapter also has isolation. Furthermore, if something breaks down when using a mass storage device, it's almost always the interface itself, not the laptop.

It's not a big effort to test for such a problem. Connect the laptop with its power adapter to the car's power supply – but without a diagnostic interface. Measure the voltage between the laptop's ground and the car's ground. No voltage? Okay. Without voltage, no current can flow. But usually, you will find some voltage. Even a good 12V to laptop power adapter might have a slight leakage current. So, measure the leakage current. Switch the multimeter to current measurement – almost all multimeters have a fuse inside, or if not, use a 1A fuse in series with the measurement. When only a few milliamperes are flowing, it's just a small leakage current and not really a problem.

Yes, the USB interfaces require 5V from the computer. We did it in a way that allows Windows to recognize the interface, so that you can install the drivers without connecting the other end to the car.

-Uwe-

Hello Roger,
It is not possible to say that with certainty based on the information provided. It could be just a 'chopper converter,' but it could also be a galvanically isolated converter.
You don't really know for sure; why don't you ask the seller?
Best regards,
Günther (formerly of gwg)

Is something like that, for example, such an unsuitable adapter ?
http://cgi.ebay.de/ws/eBayISAPI.dll?ViewItem&category=31534&item=5157711889&rd=1

I would be the least likely to trust such a cheap, generic part. This part looks very similar to a component that was sold for a short time in the USA by Radio Shack (often jokingly referred to as 'Radio Sh*t'), which caused interface problems for two customers. I wanted to buy one myself to test it out, but I can't find it anymore.

But for 26 euros, and with a '4-week right of withdrawal,' you could try it out, perform the measurements mentioned above, and if it turns out to be problematic, you can return it.

If anyone is interested in high-quality parts, I can recommend these:
http://www.caradapter.net/
However, the parts are manufactured in the USA, not in China or elsewhere. Yes, and that's why they are a bit more expensive.

This doesn't mean that there aren't cheaper options that also function well.

-Uwe-

Hello Uwe,

Thank you very much for the information. The provider also has a smaller version with 3.5 A / 70 W for only €14.90, which I had already ordered a few days ago - before I was aware of the issue of galvanic isolation . As soon as I receive my KKL interface kit from Rainer, I will simply try it out and, before connecting it, measure the voltage difference between the interface cable and the ground of the RS 232 port on the laptop. I already knew that based on the latest information I saw here.

Am I correct in assuming that this issue also applies to the serial adapter, or does it only affect the USB ?

Measuring voltage alone is not enough. I also find voltage issues even with very good power supplies. As I said, you need to determine whether this voltage is 'soft' due to a small amount of leakage current, or whether it is 'hard.'

The answer that the USB adapter needs 5V was for a different question asked earlier, where someone was trying to build a USB 'splitter' and asked if the interfaces also need 5V.

Smart interfaces have no separation, whether it's serial or USB.

-Uwe-

Hello Uwe,

Okay, I will also perform the current measurement with a fuse in series, as you described.

I didn't actually mean anything about the 5V, but I had already asked about serial adapters further up. Is everything clear now?

Thank you again for participating here .

Hi,

If you want a gentler approach, you can use a resistor instead of a fuse and then measure the current. Then, gradually decrease the resistance until you reach 0 ohms, or stop when you detect a significant current flow.

Best regards,

Jan.

Okay, the DC adapter and the kit for the diagnostic adapter have arrived. I assembled everything yesterday and put it into operation today .

The current measurement between pin 5 of the RS 232 and the notebook's ground, or the pin of the RS 232 port on the notebook, showed an absurd 3.1 mA balancing current at a voltage difference of 0.3 V. Everything is working perfectly; the DC adapter doesn't even get warm during operation. This is probably also due to the energy-efficient notebook (1.9 A @ 18 V). I just completed my first logging trip today, and it went perfectly.

I can definitely recommend this DC adapter. It's well-made, very compact, and works perfectly. When set to 18V, I measured 18.3V at the output, which is exactly the same as the original Compaq AC adapter. I can't assess the quality of the output voltage due to the lack of an oscilloscope, but it seems to be sufficient for my old Compaq.

For those who are interested, this is the item (14.90 €: [url][/url]http://cgi.ebay.de/ws/eBayISAPI.dll?ViewItem&rd=1&item=5156337635&ssPageName=STRK:MEWN:IT{MARKER}

Great!

As I mentioned earlier, a true mass closure is very rare.

-Uwe-

@Roger:

Thank you for your "detective work." I actually bought the exact same DC-DC adapter from egedöns as well. I hadn't dared to use this adapter together with the OBD interface until now.
After you have now precisely measured this specific DC-DC adapter from VANSON (model: SDR-70W), I can confidently use it.

Thank you .

Regards,
Alex.

Hi everyone,

Let's briefly discuss galvanic isolation, etc.
If a level shifter IC is used in the interface (e.g., MAX232), then there is no galvanic isolation. The standard wiring configuration for the MAX232 *always* creates a common ground between the car, the diagnostic interface, and the laptop. (Unless someone inserts a resistor into the ground line.) Well, you can also create problems through force.

Now, the question is why I can't understand the discussion about the laptop power adapter. The simple answer is: because the laptop power adapter is galvanically isolated (at least with a 230V input voltage). If the switching power supply were not galvanically isolated, it would not be VDE compliant. In this case, I wouldn't know how the manufacturer could obtain the 'CE' mark, which they need in order to be allowed to sell their product in Germany at all.

I'm puzzled by how to build a DC-DC converter (converting a 12V car voltage to a 16V laptop voltage) that has a car ground on one side and a different potential on the other. Unless someone is acting grossly negligent or intentionally, it seems the only possible explanation is that the traces are far too narrowly dimensioned. But how does such a product obtain its CE marking?

Okay, let's think this through again: The output voltage of the laptop power supply, relative to ground potential, is floating somewhere in the air. When you connect a laptop to the power supply, the laptop's ground is also floating somewhere in the air. When you connect a USB-to-RS232 adapter (which gets its power from the USB port), everything is still floating. When you connect a diagnostic interface, everything is still floating (remember: floating relative to ground potential). Now, when you connect the diagnostic interface to the car, everything is still floating. The only thing that is definitely true is that the battery negative terminal, the diagnostic interface ground, the USB-to-RS232 converter ground, the laptop ground, and the switching power supply ground are all at the same potential.

If car tires were sufficiently insulating, it would theoretically be possible to ground the car battery, connect it to a 230V outlet, or even to a 110kV high-voltage line, and everything would still function.
( Please do not try this, because at the very least, if you touch the car while another part of your body is at ground potential, it will really hurt. )


Regards,
Sure, here is the translation of the text from German to English:

'Micha'

Hi,

Quote:

I'm puzzled by how to build a DC-DC converter (converting a 12V car voltage to a 16V laptop voltage) where one side is connected to the car's ground and the other side has a different potential.

You're overlooking something... or rather, you're tacitly assuming in your considerations that the input voltage potential of the 12V-to-something power supply is identical to the ground at the RS232 port (or the computer's ground). This seems to be the case for most computers based on this thread, but it cannot be generalized unless you know how a specific laptop generates its internal voltages from the typically 16...19 volt input voltage. THAT is the problem I was talking about.

While it has never been questioned that the 230V power supplies are galvanically isolated, even in that case, the ground connection likely plays a role (with regard to interference).

My non-galvanically isolated diagnostic interface is malfunctioning when I do the following:

12 Volts -> Converter -> 230 Volts -> Laptop power adapter -> Laptop.

It works when the vehicle is stationary, but the connection is immediately lost when the engine is running. It works perfectly fine with both at a power outlet.

My (untested, even after months) assumption is that the notebook power supply should probably be grounded on the primary side (it normally has a Schuko plug) in order to bring the entire system to a defined potential. Well... I'll find some time to test that out eventually - or I'll just buy a 12V adapter.

Best regards,

Jan.

Hi,

You are tacitly assuming in your considerations that the notebook is designed in such a way that the input-side ground potential of the 12V-to-something power supply is identical to the ground at the RS232 port (or the computer ground). It seems that this is the case for most computers, according to this thread, but one cannot generalize this as long as one does not know how a specific laptop generates its internal voltages from the typically 16...19 volt input voltage.

Yes, that can be generalized, at least as long as there is a CE marking somewhere on the laptop.

12 Volts -> Converter -> 230 Volts -> Laptop Power Adapter -> Laptop
I hope that both converters aren't operating at approximately 30 kHz.

While the vehicle is stationary, it works; however, when the engine is running, the connection is immediately lost. At a power outlet, both options work perfectly.
Instead of using a DMM for measurement, try using an oscilloscope and check what your voltage looks like at the interface ICs. How exactly is the interface structured?

My (untested, even after months) assumption is that the notebook power supply should probably be grounded on the primary side (it normally has a Schuko plug) in order to bring the whole thing to a defined potential. Well... sometime I'll have the time to test that - or I'll just buy a 12V adapter.

Again: The output of the laptop power supply remains floating, regardless of which potential you connect to the primary side. I suspect there is an error in your interface. However, to provide more specific information, you would need to post or email a schematic diagram.

Regards,
Sure, here is the translation of the text from German to English:

'Micha'

@donalexo:

You're welcome .

@Bee:

The topic of ground potential is, of course, obsolete in a car due to the lack of a conductive ground connection. However, there's a similar issue here that we're discussing; in a car, it's the chassis ground potential. Basically, it's similar, but due to the generally much higher currents, it's generally more susceptible to interference. If you connect two devices at different points to the vehicle's chassis (which is almost always the case) and then connect them via a cable that also carries ground, you create a ground loop. A potential difference arises between the ground points of the two devices, which depends, among other things, on the distance from the vehicle's battery and the impedance of the connection to it. With proper current flow, for example to a power amplifier, the compensating currents can reach significant magnitudes, leading to excessive heating at the point with the highest resistance (signal cables, traces). "Wandering current flows can cause hum in radio/amplifier systems, or, in delicate circuits like those found in laptops or diagnostic adapters, they can erode the conductive traces..."

Sure, here's the translation:

"Just a quick aside:"
In the field of fault currents, I am very aware of the potential dangers. In my professional experience, I have worked extensively with -48V DC telecommunications systems, where the positive conductor is grounded. Due to internal connections within the equipment, the operating current is then divided between the positive supply line and the protective conductor. This can lead to currents in the kiloampere range (!) on the protective conductor in extensive systems and under unfavorable conditions. Consequently, the consequences regarding electromagnetic compatibility (EMC - fields) can be significant, and there is even a potential risk to personnel if work is performed on the normally current-free protective conductor. If a connection is carelessly removed in this situation, the remaining, weaker current path can quickly overload and fail. With multiple parallel runs of 300 mm² cables, a 25 mm² branch cable can easily overheat and melt. The VDE still has some catching up to do in some areas regarding corresponding regulations, which is why a colleague of mine recently joined the standardization committee there .

@Jan:

The DC adapter I purchased requires a ground connection on the outside of the DC jack. A polarity reversal is not intended - and for good reason. This should ensure that the DC ground of the laptop is connected to the vehicle's ground with a low resistance. If you then trace the ground connection of the cigarette lighter and the connection to the diagnostic port, a significant potential difference should not arise. In my opinion, this should only cause problems with laptops that have a positive voltage on the outside of their DC input jack. For vehicles where the diagnostic port is very far from the cigarette lighter (e.g., In a scenario involving a high-power laptop, such as the Phaeton, which simultaneously draws significant current through the cigarette lighter and the DC adapter cable, a potential voltage difference could arise. However, in my opinion, this difference should still be within acceptable limits and not pose a significant problem.

The reason your interface is malfunctioning with the DC-AC converter is likely due to the inherent double galvanic isolation provided by the converter and the AC power supply. This means the ground potential of the laptop is effectively floating and no longer connected to the vehicle's ground. The ground connection is only established through the diagnostic cable. This applies whether you're running directly from AC power or with the engine running. I suspect there's a fault in the diagnostic interface or its wiring, specifically a faulty ground connection somewhere that is being interrupted by the vibrations during starting.

Generally, I would advise you to get a cost-effective DC-to-AC converter after repairing the interface, mainly for the sake of better efficiency. The battery drain can become an issue with prolonged tinkering with a stationary engine. Unless, of course, you want to actively support the performance of your diesel heater with the DC-to-AC converter . My DC-DC converter remains at ambient temperature even after operating for > 60 minutes, which indicates that it has extremely low losses. I find it hard to believe that a DC-AC converter could have such minimal losses.

Hello,

Quote:

I hope that both converters aren't running at approximately 30 kHz.

I wouldn't rule it out, and in my opinion, it's also an explanation for the behavior.

Quote:

Instead of using a DMM for measurement, try using an oscilloscope and check what your voltage looks like at the interface ICs. How exactly is the interface structured?

It's a purchased interface (not from here... Rainer's shop didn't even exist when I bought it), and it doesn't have galvanic isolation, which means it can't be opened without risking damage. I want to reiterate that it also works perfectly fine with the engine running, whether the laptop is powered by the battery (which is no longer possible) or by the power adapter - using the same power adapter.

Quote:

Again: The output of the laptop power supply remains floating, regardless of which potential you connect to the primary side.

"I understand that perfectly. My goal is to reduce interference, which I believe is one of the possible causes. There's probably a reason why IBM equips the power supply with a Schuko plug instead of just using Euro plugs without grounding, like most other manufacturers. My hope is that it will improve electromagnetic compatibility (EMC) if I ensure that the "power source" (i.e., the car + converter) and the power supply have the same ground potential." It's clear that this won't go through to the secondary page.

Quote:

I suspect there is a bug in your interface.

I can't imagine any interface error that would only occur when powering the notebook directly from a 12->230V AC adapter without any grounding, and while the motor is running simultaneously. The motor functions properly when running on a "real" mains power supply.

Quote:

The DC adapter I purchased requires a specific grounding configuration on the outside of the DC jack.

At the very least, that's compatible with my IBM machine.

Quote:

The chassis ground potential of the notebook is therefore floating and no longer referenced to the vehicle's chassis ground. The ground connection is established solely through the diagnostic cable. This applies both when operating directly from AC power and when the motor is running.

"As you said... this also applies when operating with direct AC power... if the issue isn't due to electromagnetic compatibility (EMC) caused by the 'double galvanic isolation,' then the exact same problems should also occur when operating with AC power."

Quote:

I therefore suspect a fault in the diagnostic interface or its wiring, specifically a faulty ground connection somewhere, which is interrupted by the jolts during starting.

"As mentioned above, if that were the case, it would also malfunction in AC operation. It has never done so, and it has undergone numerous startup procedures without any subsequent issues." Problems only occurred during the tests conducted on the converter that were performed between the initial and final stages.

Quote:

Basically, after the interface is repaired, I would recommend a cost-effective DC-DC converter.

I've also been thinking about that (and probably will continue to do so), especially considering the prices and the fact that it would mean one less component in the car that you need. For me, energy consumption is less of a concern because I already have AC power available for my crafting, and the maximum 10 amps doesn't really matter when driving – it's probably going to be less than that anyway.

I will leave the "repair" interface as is... because, for the reasons mentioned above, I consider an error to be very unlikely.

Best regards,

Jan.