2022 12 15 HW

Okay.
Shall we start or do we need to wait?
I think we can start, yeah.
We wanted to continue with the schematic review actually, is the first topic.
Or, I don't know.
We can shift it afterwards, after we go through the points.
This is also happening.
This is also happening, yeah.
Yeah, because if for example Axel will join, because he informed us he will also join, then we can directly start.
Let me just check, because here regarding this topic, we actually need Axel.
Maybe we can discuss at first about the EMC topics and then jump back to the DTC issue with the circuit breaker temperature.
Okay.
Yeah, I guess this will be my topic.
So summary is difficult to prepare at this time, because we are waiting for several reports.
Because we did a lot of external tests now at KFE and they are now trying to finish all the reports before the Christmas vacation.
But maybe I will not make it until next week, but then it will be finished in January, early January.
What we did this week was another emission test with CP to check the latest changes of the CPIC inverter.
As you saw in the schematics review, one diode will be taken out of this.
And this is the modification we have tested by connecting it with two different variants with a little solder and a lot of solder.
And there is no change in the emission.
So this is a good result. So we can develop this diode and are changing the layout accordingly.
And then the other topic what we want to confirm is the BCI behavior with the changes.
What we are planning to do on the cell ASIC.
But this test is not finally conclusive, because also partly cold temperatures and lack of time then to complete it.
And we are now trying to get internal slots in discussion with our EMC lab.
So for this, we cannot give a final message now.
Understood. Okay.
So this means that most probably we will get some results beginning of next year in January sometime.
Yep. This is my expectation.
Okay. Understood.
Okay. Thank you. Maybe also another organizational topic.
I think from the BMW point of view, I think this week will be also the last meeting from the hardware.
I don't know if from Hella side there are still persons working next week.
But I think that in our case both Karina and I will be out of office starting next week.
Yeah.
What about the new year? Do we have there also?
Do you mean the first week?
Yeah, for example.
So we will start or I will start at least from 9th of January.
So this means that basically, yeah, on 5th of January, we will not go or I will not be in the office.
I don't know, Karina, are you in the office on 5th of January?
No, I'm not in the office as well.
So this means we can cancel next week's meeting and the meeting from the 5th of January basically.
Okay. Thanks.
Just as an information for you. Okay?
Yeah. Thank you.
Okay. I see Axel. I see Jan.
So I think it makes sense to start with the temperature topic.
Yeah, of course. Hello.
Hi. I would take over this.
I imagine it's about Karina, your email from yesterday. No?
Yeah, I think we can start with this email topic. So maybe just for the others, you are also informed.
I asked you to get some insight about the test results where we saw the 28 degree difference between the two NTCs.
Just that we can see there's really this huge difference and how to go on with the topic because up to now we have not seen the results.
Yeah. The problem is that this results, I will not show something.
Okay. First of all, the problem is I explained in the meeting, in our thermal meeting, that we damaged this thermal sample.
Unfortunately, this test where we have seen this huge difference was the last one before we have seen that it was damaged.
Yeah. We analyzed everything and the behavior is strange. Yeah.
That's why we cannot, how to say, we cannot guarantee that this is the sample itself was in a good state as we did this test.
Yeah. And that's why we will reduce everything. Yeah. Heat cycle B is already done.
I have presented the results on Monday. Yeah.
The results, the difference was around about eight, 10 degrees C. So in the same range we had in the past from the measured temperatures on the
on the thermocouples. Do not exactly remember the value, but it was not that huge.
It's not the test where we see these big differences. Yeah.
Because the big differences, we see it in so-called daily profiles. We have this highest test defined where we have 425 amps for 10 seconds, for example.
Yeah. This pulse itself is what results in bigger deviations. But as I said, due to the fact that the sample was damaged right afterwards.
Yeah. We cannot rely on these results.
What we have explained in this thermal meeting is that the idea behind this monitoring is not to catch the dynamic behavior as to see that it is continuously damaged.
Yeah. Means that if we have for a short period of time, the high differences, what we see in the test and what is explainable due to the position of the thermocouples of the disease.
Yeah. We simply so it's not the right way to go only via threshold itself. Yeah.
We have to increase the debounce time and at least Monday, what we have discussed the proposal or the way the direction would be to have something around about 10 degrees C as a threshold and then round about two minutes of the bouncing.
Yeah. To finally set these thresholds, these values, we have to wait until everything is finalized.
Yeah. I said with this new sample, we have heat cycle B behavior. We have already done the test.
Yes. Simply to have the statement that it works, let's say. Yeah. But we have to redo a lot and this is ongoing.
Okay. So we also discussed again internally about the debouncing time if this is the right solution, but up to now, we are not really happy just increasing the debounce time.
We really want to understand why it's not possible. Malati, I can hear you. Can you maybe?
Yeah. Okay. Now it works. Okay. Yeah. We really want to understand why it's not possible to change the placing of the NTC because from our understanding, just changing only the layout, it is quite easy.
So we want to see what's the background of your decision not changing the NTC. Understood. Yeah. We have discussed this as well. Yeah. We have checked it is in general possible.
At the moment, we are in this C2 design phase where we do not touch the power PCB. Yeah. So if we would move this one NTC, this would be the only change which is at the moment not fully clear if it will have the expected results.
Yeah, it will improve a bit. It will reduce the difference. Yeah. But in our understanding what we have seen, we can detect what we want to detect.
Let's say. Okay. So my major question is, of course, the plausibilization of the two NTCs. This is a kind of functional safety requirement coming from the functional safety point of view.
So is it, let's say from Christoph, my justice man, is it totally fine to deal with that debouncing time of two minutes? It is. Yeah. Because there is a set.
Yeah. The only reason for this check is to see that something is damaged forever. Let's say. Yeah. Okay. So it's not necessary for my functional safety point of view to have a really fast feedback for, let's say, to have all the time and temperature
plausibilization to be safe according short circuit or, let's say, any high current pulse with, let's say, 600 amps for one second or the, let's say, 680 amps for 0.5 seconds. So this is not linked to that. It's only for total damage of the VMS.
Is that correct?
This is what we have discussed. Yeah. I can try to ask it differently and formulate it as you just formulated. And I will check it once again. But this is what we have discussed. Yes.
Okay. So this is, let's say, our major concern.
And it's just that you understand why we are pushing on that topic. And this is, let's say, definitely only, let's say, that we have here robust design, not that we have, let's say, a kind of temperature plausibilization then with a kind of buffer time. And then at the end, we have a problem with the short circuit in some use cases.
And then at the end, let's say, for example, the TIF suit and functional safety assessment say, oh, that's not okay. And that is not according to an ASO C requirement for the self protection of the battery.
This is, let's say, my major concern. And just regarding that, and if you can confirm it by Christoph and perhaps also already, let's say, by first discussion with the colleagues from TIF suit, I do not know if you have the possibility to check that.
And also having a little checkup with Andreas, that would be quite good to have a short, short meeting and just to have an hour about that. And then we are fine. So when this is confirmed, we don't need to change the entities.
But if there are, let's say, concerns or open topics, then I just want to have the other way around, say, okay, a new position of the entity would be good, I think, from the whole point of view, from the whole concept and so on looks much better on that other side.
And also in our point of view, we totally understand that you don't want to change anymore the power PCP. Change itself was quite low or small. Sorry. In our point of view, of course, you have, let's say, redesign, you have reprogramming of the production and so on.
So there are major steps. But on the other hand, if I'm comparing to that how often we are discussing that right now with you in that meeting, it's 12 people. This is also much money.
We need to we need to find the correct solution. And that's, let's say, what I just want to point out. Yeah. Yeah. I have a meeting today with Christopher and about for me. Yeah, I will. Great. Great. In this meeting once again.
But the set. Yeah. So we discussed this internally and it was fine. Yeah. Yeah. Then it's also fine for us when we have a robust solution that was, let's say, 10 degrees difference and kind of buffer time. I do not know which one.
Of course, some some seconds or let's say one minute. I do not know. I think I heard two minutes. So this is something we need to discuss. Perhaps there's also another solution that we say, OK, we comply with the current, which was within the last, I do not know, 10 seconds applied to the circuit.
Like it could be also a possibility. I think there are solutions to to respect the locations of the entities right now. But I just want to be sure that we have a correct AESL C according to that self protection.
Okay. Yeah. Great. Good. And let's just just give us a feedback written feedback by by mail. When you when you talk with Christopher Christopher and that's great.
Thanks a lot.
Thank you. Also from our side. I think we can jump to the next topics. Right. Yeah. Yeah.
Sorry, Malati.
Yeah. Shall we discuss about this topic? Second source components.
And here will be the question would be what's the status internally, because as far as I understood, you are still trying to procure the on semi MOSFETs.
I don't know if you have any updates on this.
We do not have updates.
We ordered some parts on our sample shop to get PCBs power PCBs with this new.
Okay. It's ongoing.
Okay.
We defined also variance here, which we want to get.
What exactly do you mean with variance?
Yeah. How much how much PCBs with which amount of transistors and so on.
Okay. This is what you meant. Okay.
Hello.
Yes, we can hear you.
I have some network issues.
You have your points.
It's the discuss with you via email.
Okay.
Yeah, because let's say you could not attend the last meeting. Okay. And this was the idea behind.
This is one topic where we need some information from your side.
Have you written an email?
I haven't read or write.
What is this? Yeah, I thought we have provided everything to you.
You wanted to check it.
This is my state. Is it not correct?
I have to check the error metrics.
Yeah, because in the error metrics, it's mentioned what is checked in the oil and whatnot.
And you wanted to check our accuracy.
The topic is clarified. This is clarified. I think it's also close in the meantime.
Did you already do the update of the error metrics? Because I think this is the information. What's what's missing?
It is on Panama. Yeah, it is on Panama.
Okay, because the latest one last week.
Okay, so there is an update of the error metrics because this is what I was waiting for. Okay.
Understood.
Let me check the latest update. It's indeed from 5th of December. Okay.
Okay. Yeah. Okay. Then this is a to do topic for for me.
I will check it. Perfect.
The next topic is technical product description regarding the requirements specification.
So, Julia, did you export it completely?
We had a discussion with Jan Moritz and my proposal was strange to talk.
Okay. We had the discussion. My proposal was not a good one, let's say.
Yeah, we have filled the template they provided yesterday.
Moritz has written an email to me with the review findings.
So I have to update it, but there is already one available on Panama.
Do you want to have the link because it's in the chapter?
Yeah, if you can put the link in the chat would be nice.
Yes, I completely right. We are working on that together with Moritz and we just wrote you yesterday an email with just missing information.
Julia, exactly what you mentioned. So when you send it to us, we will check it once again, especially Moritz.
And then let's say if there's anything which needs to be optimized, then we will come back to you.
Otherwise, if it is fine so far from BMS point of view and from functional point of view, we will provide it further on to EVE side so that EVE can fill out all the relevant information.
And then let's say we will also make a make a loop once again for checking. But I think that's continuously ongoing.
So if anyone's interested, I think it's good that you just shared the link for us. But anyway, we're working on that.
Just did it in the chat.
Thanks.
So Zorin, if you get to know if there's any hard deadline or something like that, let me know.
I would not be aware of this. So that's just continuously ongoing until we finish it.
No, nine.
Okay. So Jan is fine if I will not do it this week. Yeah, because.
Of course.
Yeah, this issue. So we have a plan and we have already offered everything.
The problem is that in the software, it needs to be put on priority.
I think, Karina, again to you, I think you asked my email about it and I replied in the email.
And I stated that from my understanding, the earliest time point we can bring it is three ninety.
Yeah, because three fifty. Please wait.
It's no problem. It's important for us.
It's just important that we know that you are or that the hardware is communicating with the software that we are well when you will implement it.
Yeah, to keep it in mind that we have kind of write down what we need in the software, what we shifted from hardware to software that and that it's not in the end.
Yeah, but three ninety is most likely not possible from the software point of view.
Yeah, because I said three fifty is already more or less done. Yeah, three seventy. There is only one month between three fifty and three seventy.
Yeah, the main focus there will be on functional safety because we do not have to.
Yeah, we are forced to put higher maturity levels and yeah, the progress is not in that way.
Three ninety would be the very, very first one. Yeah. But after the discussion with Zimmon yesterday,
if we will not say this is the highest priority, then it will not come in the three ninety.
Yeah, I think we only what we have to keep in mind is that we have a few ground loss tests in the PV.
And I think for these tests we need we need the software implementation.
Just about also to say, yeah, maybe you're not able to implement this directly in an official ice step.
But is it possible to get at least a test of that just to make sure that the fix will work?
Because sometimes if I take a look at this, can a recovery fail during a test?
It might be that software, let's say, fix could hide another, I don't know, hardware issue.
And this is something we would like to avoid. Yeah, I know.
I know from Valentin that it works. Right.
So it was already proven in lab test.
But the problem is just that we needed for the 12 volt or 48 volt sequence because there we have a load down or a ground loss.
And the timeline, let me check.
Maybe we can state it here. Yeah.
I have to open the PV plan.
Okay.
So the test is planned for.
June. Yeah. Okay. Maybe it is a comment.
Yeah. This mid of June, because this would be then the absolute last deadline.
Let's say.
So maybe this is for the for this issue for the ground loss detection.
But the topic itself, we call it the status of the implementation.
Actually, there are two additional things which we would like to track with this topic.
And as you can read there, you have this new change with the sample with this additional path for the 12 volt supply in order to have a faster reaction.
As far as I understood, this needs also a softer change.
And of course, you have to the topic with the recovery.
Yeah. The second one is easier. Yeah. The second one, this additional pass for 12.
Yeah, this is a kind of configuration matter.
And yeah, it will.
It does not have complete complex implementation behind.
You have to configure it correctly and you have to react on it.
Yeah, this I can double check, but this will be easier to bring it in this other issue.
Yeah.
We are in analysis with Valentin. Yeah.
In my early to said that there is a software implementation needed.
Yeah. First needs to find out why. Yeah.
Yeah. So still no no identified root cause for this topic.
In the next meeting, yesterday we created the test setup. Yeah.
So that we can reproduce it in development now. Yeah.
And from today on, we can then analyze what exactly happens.
Can there be an issue between switching off and changing the power supply of the RX to the B plus clamp?
Because I think this is kind of what happens.
And so we are losing clamp 30, then the microcontroller or the P mix switches the 12 volt safety power supply and gets the power from the B plus.
And then we have kind of a different supply during a short time.
Yeah, sure. Yeah.
This is a kind of critical path here. Yeah.
But give us please the time. Yeah.
I do not know. You are the complete week out of office.
I mean, Valentin, you are not in the office next week as well.
Maybe we can write an email when we have the analysis result because we wanted to do it today. Yeah.
Yeah. Yeah. Of course. Please write an email. I will also check my emails next week.
OK. Yeah. Because speculation is not always good. Yeah.
Because. Yeah. Yeah. Of course. So just only a loose of clamp 30. It's working fine.
There is different other behaviors.
So fast. So very slow rising of claims. It's maybe cause. Yeah.
Like or something like that. That we are there. We are in another analysis.
I cannot explain you the detail behavior. I have to bring there more visibility from technical point of view.
Yeah. Of course. OK. Yeah. When can you are able to provide the report to you?
Which reports. You can inform after the analysis.
We will. Exactly. We will write an email latest tomorrow because next week Valentin is not in the office.
So today and tomorrow the setup is not fully blocked here because we needed for other tests as well.
But. One two hours today and maybe one two hours tomorrow.

 

Which reports?
We can inform after the analysis.
We will write an email, latest tomorrow, because next week Warlinden is not in the office.
So today and tomorrow the setup is not fully blocked, because we need it for other tests as well.
But one, two hours today and maybe one, two hours tomorrow.
I think Malati, this is the other point about the software implementation.
I think you have to go back on the agenda.
It's the last one.
So here you can write down for the CanRecovery fail that analysis is ongoing and we will get feedback via email maybe next week.
Okay.
So any update on this one? Design review checkups?
I think we have shifted this topic for the beginning of next year.
I think now that we just discussed, actually we need to shift it to not 5th, but to 12th of January.
Because we need clarification on functional safety side.
Exactly, we wanted to discuss internally with our colleague Andreas, but he was yesterday and the day before he was out of the office.
That's why we could not reach him.
To be honest, I don't think this is also a priority topic, so I think it's fine.
Okay, this is the load dump test.
So about the load dump test, I got your email Julia.
So Julia provided some informational questions to EBE, which has to be clarified, whether to check if the battery is damaged or not.
That's still an open question. And in parallel, we requested to restart a test with the lower inductance, about 10 microhenry with a C1 sample.
Just so don't lose more time with the load dump test. I think this is an issue which is ongoing since 4 months and still not solved on the EBE side.
What is the inductance value, Karina?
10 microhenry.
Sorry, does it make sense to send us this BMS from the load dump battery then? Because in case we don't have their progress, maybe we can somehow check it on our side.
I think that will be faster.
Actually, we said that it's faster if they read out everything before they open. Opening of the battery is not that easy.
That's why first read out everything and then.
Okay, yeah, this was.
This is because I lost the network connection. It's discussed already. Yes, we have discussed this.
You have topics for me. Can you please check because I would like to see.
Thank you and I don't think we will hear each other anymore. You know, wish you a Merry Christmas. Good year.
All the best and hear from you next year.
Oh, she's out.
Thank you.
You can pass it.
I can forward.
So we have sent to go into the schematic review again, right?
Yes.
I will share my screen.
So last day, so we updated the schematic with some changes, minor changes.
I will point just only this changes. We changed the reference.
I think let me help.
I think I think where we remained last time was at M 38. This is the cell isolator chip.
Yes, yes, yes, I know. I know this.
But I wanted to point out that we provided here.
We rechanged our changes back, I will say, because of some hints from tests.
And therefore I will show you that we, for example, for this reference on microcontroller, we changed it back because we.
So basically you reintroduced the Xena diode to back.
So there will be no change, basically.
Yes, yes, we have here a parallel path.
It stays, but it is disconnected, but not populating off this resistor.
OK, that's interesting.
We will be the exact reason behind it.
But found out that clarification and also some feelings that maybe we will get problems in MC.
OK, but in case we nothing change here, then the MC will be the same as before.
So therefore it's just only changed back.
Not more.
This point.
And on.
The schematic page for the.
Cell isolator, we are in clarification.
This filtering here on but been sufficient or not, so the frequency corner that we have to.
Yeah, somehow to found out.
We planning here some tests and yeah, to keep the hot plug solved and to get here the right values.
Yeah, we are.
Yeah.
And the investigation, I will say ongoing investigation.
OK, then we.
Go to this isolator here.
Nothing changed.
Nothing changed.
This means you will not do any changes.
Yes, OK, from C1 to C2.
We planned here to to check the saving of this.
I will say component because we have here just only functional isolation.
What I explained maybe last time, but.
In case of some doubts, we don't change this point here.
It's also good news.
Actually, yes, never change a running system.
OK, good.
So then we we are here on the over current protection.
Here we have two changes.
OK, mainly the input resistors of inner preamplifiers.
We increased from one ohm to ten ohms.
To get here.
Better or to to match the better simulation results with simulating here all things and to get here more robustness.
The input input.
Now we decided to increase this resistor.
It is from application point of view.
OK, fine.
OK.
This will change a little bit basically the the the the the current.
Yeah, of course.
And also the frequency.
Yeah, it's as dynamic, but it is it is fine also to react fast here for work on this.
It's fine.
It's not a problem.
Will we see any changes in the measurement accuracy or this will not be impacted?
No, that is not being affected.
OK.
OK, and we changed here.
The dynamic reaction of over current.
You see here that seven hundred seventy amps are static in positive current or in battery deloading direction.
We can have here static currents up to seven hundred seventy amps.
But dynamically, we react somehow with yeah, with faster.
So faster timing, I will say.
So we see here one thousand one hundred and matching to one point five millisecond.
OK, but the actual the change itself, it's you've changed the low pass filter there.
Yes, we changed here.
Only is this two values to match this somehow requirement before it was some more for some more time possible.
But we reduced somehow to yeah, to.
Yeah, to to to protect to be protected somehow better in the blind spot where the software cannot react.
I guess my next question, do we see an impact due to this change in the software?
Are there any thresholds which need to be adjusted due to the new reaction behavior?
Yeah, what we got from software side that in 10 milliseconds somehow it's not really possible to react by software.
And that is the definition of blind spot.
And sure, the timings here on lower current are much higher.
You know, and therefore we react somehow.
Yeah, faster as before, because we clarified this also point with functional safety and also Julia.
And it is fine to do so also on requirement level.
We have to somehow to protect us.
And that is here, I will say improved.
Additionally, but we also have here for, for example, for the short circuit protection,
also the other threshold up to one thousand one hundred fifty amps, right?
Because here below, it's a dynamic and dynamical behavior.
And here above, it's a static.
So in case the current is very fast rising, very.
Yeah, then we can somehow handle this statically.
One question. What was the reaction time before, before the change for this two values, one thousand two hundred and one thousand amps?
Here we were roughly on, I think, five milliseconds here on one thousand.
We were one hundred milliseconds.
So so very, very more time we had here.
But in case, yeah, in case of I will say sequence scenario, in case we add all sequence after was all pulses.
Then it's all this to high impact to the circuit breaker and clarify this.
These values are fine from requirement point of view on system level.
So for us, this means basically that the hardware will now react much faster to this to this current peaks on overcurrent.
Yeah, exactly.
On overcurrent. OK, thank you.
Is this OK?
And what what I do mean is basically since before the direction was.
But yeah, this type of overcurrents could happen, but they were not detected if they were under, let's say, one hundred milliseconds.
Yeah, for one thousand amps, as you said, and now it's way faster.
I think from my understanding, the one thousand two hundred amps should be shut off after one millisecond.
So this was the requirement.
So since it was not possible in software, this should be a good solution.
Now doing it in hardware because the requirement says one millisecond for one thousand two hundred amps.
I also read I will also check it again.
But this was my understanding.
I don't have the values in mind.
The value was the last value from BMW was to match or to conduct six hundred eighty amps anyway.
So and not react from hardware.
Yeah, and that is definitely the case because we react statically from seven hundred seventy amps on and all over.
All currents should be matched or should be caged from software.
And so, OK, and that is how fast is then the overcurrent protection for the one thousand two hundred fifty amps?
And which microseconds range that I can show you?
Yeah, good question.
Maybe I can show it directly from test result that can provide you a better feeling.
I also. I need some seconds.
It's around about 50 microseconds.
I'll show you that. It's. Because we have a bit less, but we have somehow to interpret this curve.
We see we apply here on the left side here the current in orange here.
And it's rising up somehow up to this.
I will say value that is 200 and for this.
So one, two, three, four, five, six, it's around about one thousand two hundred fifty amps.
And the reaction is here happened.
You see the CB on signal is going down.
Yeah. And from here, from this point for disconnecting that is a very long year, we have just only less than 20 microseconds.
It is a short circuit disconnection.
It's very fast.
That is a static disconnection behavior.
OK, so we have roughly less than 20 or around 20 microseconds reaction time.
Yeah, around 20 microseconds.
So normally the disconnection is here done here because this is the current which is stopped, I will say.
And then we have free free free oscillating in the system because capacitance and inductivity is given in the system.
But I think below 20 microseconds here, the hardware reaction is given.
And this timing before it's rising up, the current, we see it's very fast rising and very fast disconnecting.
OK, thank you.
That is the C1 sample.
That's not standing here, but OK.
OK, that is only two changes here on this schematic.
Sure, we summarized some of the changes and will forward to you, which last meeting was requested.
That presentation. OK, sounds good.
But we have to update this and I think this end of this week we can provide it.
Sounds good. If we can send it end of this week, we will check it when we will get back.
I don't know, Karina, you are writing beginning of next week, still one or two days in the office.
But yes, we will definitely get the feedback from our side.
Here we see the same component value, so the reaction will be the same somehow from the redundant path.
I just wanted to say this is the redundant path. OK.
Yes. OK, then here we had just only one change for the package.
We changed here this from the current Miro here circuitry.
We changed just only this package from 0402 to 0603.
It's coming from module specification.
Found out that we have to increase the package because here we will have continuous current somehow.
And we have to ensure that power dissipation of each component is fine.
And also this PMW scaling, this factors, scale factors are also ensured here.
For power dissipation, for temperatures and so on.
Yes, that is the only change here.
Here on the interface to the power, we had here a change for the diagnosis of TVS diodes.
So before it was just only one detectable.
And now from system requested, we can detect two TVS diodes damaged in the path.
First of all, from my understanding, where are these protection diodes implemented?
Just a moment. I will open the power and show you that.
So the TVS diodes are located here in the protection circuitry in between clamp 40 and clamp 41 here.
We have five of these and this signal goes to logic and will be analyzed on logic side.
Just check for a short circuit and if you measure.
Yes, yes, yes.
Low path, okay.
In case we have here lower value, then some diodes should be then go in stuck at close mode, right?
And only for this diagnosis, for multiple point diagnosis, the system requested and we implemented this different, I will say here.
We increased this BIOS current somehow to have more spread in the diagnosis and we changed the somehow the scaling here.
What exactly did you change? Just value changes?
Yes, we changed this value and this value to get the right diagnosis.
I will say visibility in this five-fold signal, not more.
Understood.
So, yeah, and additionally, normally it's all changes, but we have here founded out by motor specification that we need here more.
I will say area provided for the cooling of this.
The history of this module is that we used before LDO, which is not available on the market, and we exchanged it by discrete circuitry here with the reference here to be independent of semiconductor market.
And therefore, we need for this transistor better cooling and we.
So just increase in the layout?
Yes, yes, yes.
Okay.
Just only here the hint and on the collector, I think, we provided there in two layers, somehow the thermal conducting.
What package does the transistor have?
Two to three, easiest or smallest package, therefore, we need here because we have here more or less five volt output, 13 voltage input.
And okay, so just a small LDO, the Xena diode.
Okay, just maximum, I think, current of 10 milliamps. It's not a lot, but for the to be here automotive, we have here to ensure the best cooling or the required cooling for this transistor.
That's all.
Okay, that's all.
Now that you just presented us this topic, have you checked the both the logic PCB as well as the power PCB if you have any additional hotspots?
Because I think this, let's say, change here comes after doing some measurements internally and you have seen okay the temperature of the.
No, this topic come up by model specification because in model specification, we calculating with maximum, for example, used currents.
For example, for this, there is in the, just a moment, I'll show you that in the schematic, for example, we're using this independent somehow here.
We don't see your screen anymore.
Open another file.
So in case we are using five volt independent.
So here and we can get some cases where we have statically more consumption current, but that it was not visible on thermal investigation.
So therefore, this improvement on logic layout.
Just a moment.
In case, for example, we provide here the test.
Yeah, then we have to get here more currents flow.
Oh, that resistor.
Because we provide here the short to that to provide here the testability.
So this is.
OK, so it is that increase.
I will say the load statically and that was not visible on the termination question investigation.
OK, now I understand.
But normally this this check, this overcurrent check shouldn't take too much time.
How much time do you keep this impinges activated in order to check if there's a short circuit?
Yes, but for the calculation, we need to exceed.
OK, we need to calculate with the static value there because we don't know how long time the software will be useful.
And to ensure that the software just only using one second, I will say it's OK.
If it's not possible, I will say it's not visible.
That is one example, and therefore we decided to increase cooling area for the system.
But I wanted to show maybe the.
Temography results for one. To we checked also the power specie and logic, the tomography and to find out the right hotspots.
And so that's yeah, I can show something, but I have to search.
But in general, we are ready with schematic review and I wanted to provide you this week and updated presentation with this hardware changes.
And then you can have an imagination.
What we will change.
We have additional questions.
Right in this moment, I have no further question. Thanks.
But yeah, please, if you managed to to finalize this presentation by end of this week, send it to us.
If not, then beginning of next week or something and we will double check it.
We will be back in the office.
Yes, that's fine. Then I would say thank you. Thank you a lot from my side.
Yes, thank you. Thank you.
Thank you too.
Now we can we can congratulate and to wish everybody a Merry Christmas and.
Thank you. Happy New Year.
Merry Christmas. Thank you.
Thank you for you as well. All the best. Have a nice night.
See you again next year.
OK.
Bye bye.
Thank you. Bye bye. Bye bye.