rusefi.com

Do we recognize this 5 plug connector to know a simple junk yard source of vehicle/harness side of ADV? Any other sources of vehicle/harness side of ADV connector?

mck1117 wrote: ↑

Thu Apr 07, 2022 8:34 pm

Bosch trims the sensor element itself on the LSU ADV. It works the same way, except there's no extra external (in the connector) resistor paralleled with the sense resistor inside the controller.

OK ..
precise laser cutter can do this job ....
..and the we can have same sensors ...

JRD McLAREN wrote: ↑

Thu Apr 07, 2022 8:37 pm

mck1117 wrote: ↑

Thu Apr 07, 2022 8:34 pm

Bosch trims the sensor element itself on the LSU ADV. It works the same way, except there's no extra external (in the connector) resistor paralleled with the sense resistor inside the controller.

OK ..
precise laser cutter can do this job ....
..and the we can have same sensors ...

no, that's only possible because of the more precise manufacturing tolerances of the actual sensor element. I don't know that they actually trim it, they may have the production well controlled such that no trim is necessary.

@dron0gus suggests 1J0973705 as the keyword for 5 pin

F 02U B00 937-01 https://www.ecotrons.com/files/Bosch%20LSU%20ADV%20sensor.pdf
https://webshop.swindonpowertrain.com/index.php?route=product/product&product_id=229

Drain and Source should be swapped? Or it is not an overvoltage protection? Just a reverse-protection?

That's for reverse protection without the drop of a diode. The body diode on Q1 "kick starts" the circuit that pulls its gate low, turning it on fully.

edit: the D4 TVS is for overvoltage protection

mck1117 wrote: ↑

Sat Apr 09, 2022 8:53 am

That's for reverse protection without the drop of a diode. The body diode on Q1 "kick starts" the circuit that pulls its gate low, turning it on fully.

edit: the D4 TVS is for overvoltage protection

Thanks for clarifying. This rises two new questions
1. Do we need to pass +12 for heater through controller? May be it should be connected to +12 outside controller? Through fuse and relay, as user wish. We can still protect heater from overvoltage by measuring Vbatt and disabling heater low side switch.
2. If we decide to connect Heater+ to +12 outside block, can we replace this reverse voltage protection circuit with simple diode?

Updated F103 wideband. Reviewers are welcomed.

rev2 ordered

looks like this preview mixes Metric and Imperial systems?

Dron_Gus wrote: ↑

Sat Apr 09, 2022 8:33 pm

Updated F103 wideband. Reviewers are welcomed.

I have checked the schematic, and I see that there is a lot of changes compared with f042 board.

Why not keep both boards similar into the LSU controller circuit?

joaodh wrote: ↑

Wed Apr 13, 2022 1:31 pm

I have checked the schematic, and I see that there is a lot of changes compared with f042 board.

Why not keep both boards similar into the LSU controller circuit?

We want to support LSU4.2 and ADV also.
We planing to add some SW diagnostic.
So some changes were added to support these.

..and two outs are planed for 0-5V OUT and 0-1V OUT
a.k.a. Lambda out .. ??- 1Volt = Lambda 1 (0.85V = 0.85Lam)
..or Narowband output ..?

JRD McLAREN wrote: ↑

Wed Apr 13, 2022 6:54 pm

..and two outs are planed for 0-5V OUT and 0-1V OUT
a.k.a. Lambda out .. ??- 1Volt = Lambda 1 (0.85V = 0.85Lam)
..or Narowband output ..?

Both outputs are capable to output in 0..5V range.
You can define any output voltage vs AFR curve.
Currently no curve defined

...cause .. this is the cheapest Lambda display ... https://www.ebay.co.uk/itm/263849308282?hash=item3d6ea4f07a:g:HPoAAOSw0xhg5vGj

.. if we can have programed Out ... 1V = Lambda 1

JRD McLAREN wrote: ↑

Wed Apr 13, 2022 7:24 pm

...cause .. this is the cheapest Lambda display ... https://www.ebay.co.uk/itm/263849308282

Thank you that's a great idea!!!

Niet zač ...

(it works on EGT also ... AD595)

Dron_Gus wrote: ↑

Wed Apr 13, 2022 5:39 pm

We want to support LSU4.2 and ADV also.
We planing to add some SW diagnostic.
So some changes were added to support these.

Ok, this is great support multiple sensors.
Regarding the schematic I have some doubts:
1- Do we need two virtual grounds?
2- Why R41, R42, R25 it's need?
3- The U1D it's expected to be a buffer, why the negative(13) it's not connect to output(14)?
4- What is the function of heater sense?
5- Why the Nernst_4.9_bias it's connected to MCU? It's to don't apply any current during startup?
6- The diode D6 only support 150mA I think we should use one with some more current (ex: SS16, LCSC: C2481)
7 - Why not place the Vbatt_sense on safe side after diode? we just want a rough measure from the battery right?

joaodh wrote: ↑

Sat Apr 16, 2022 12:02 pm

Ok, this is great support multiple sensors.
Regarding the schematic I have some doubts:
1- Do we need two virtual grounds?
2- Why R41, R42, R25 it's need?
3- The U1D it's expected to be a buffer, why the negative(13) it's not connect to output(14)?
4- What is the function of heater sense?
5- Why the Nernst_4.9_bias it's connected to MCU? It's to don't apply any current during startup?
6- The diode D6 only support 150mA I think we should use one with some more current (ex: SS16, LCSC: C2481)
7 - Why not place the Vbatt_sense on safe side after diode? we just want a rough measure from the battery right?

Thanks for review. My comments below.
1. No. But I'd like to keep two separate: one that goes outside and can be affected by external noise and one internal that is not affected and clean.
2. R41, R25 - current limiting, for protection in case of short to +12. Yes, I know that U1 will not survive if LSU_Vm or LSU_Ip or LSU_Rtrim is shorted to +12...
R42 is just for bias current balancing. "Good practice" nothing more. https://e2e.ti.com/blogs_/archives/b/thesignal/posts/input-bias-current-cancelation-resistors-do-you-really-need-them
3. It is. Through R41 (see above) and current limiting R5.
4. To check chat heater/wiring is ok. Voltage should rise to Vbat when mosfet is off. If not - there is break in circuit.
5. We need this biasing current only for LSU4.9 and LSU_ADV, and should disable for 4.2. So it is gpio driven.
6. Only MCU and OpAmp supplied through this diode. But not a sensor heater. I think 150mA should be enough. I'll double check.
7. Agree. Actually we can remove this sense and use only heater sense - it will show battery voltage while mosfet is off.

I'm a little concerned about this resistors. Be aware that they can also cause some instability on circuits.[url]https://e2e.ti.com/blogs_/archives/b/precisionhub/posts/resistors-in-the-feedback-of-a-buffer-ask-why[/l]

Do you already bring up the new board? At the moment I have one pcb from f042 with some wires connected to a bluepill board . But at the moment the HW it's already to much different

Progress

maybe it's too late, I have no idea how many pcbs you ordered ...

but is possible to modify PIN ORDER on main connector
to match the pin order on LSU sensor .. ??

somethink like:
1*-LSU IP - LSU Vm -*7
2*-Heat- - Heat+ -*8
3*-LSU Rt - LSU Un-*9
4*-Vout1 - Vout2-*10
5*-CanL - CanH-*11
6*-GND - 12Raw-*12

I was thinking about it. But...
Heater in between of 2 and 2 signals is not good for routing.
So heater was moved closer to 12V and GND. This also shortest high current routes.

1A is not high curent ...
look at my addon board ..

It will be better for users .. have wire order one-to-one ...

EDITED:
it have 4 layers ..? (pcb)

JRD McLAREN wrote: ↑

Fri Apr 22, 2022 2:05 pm

1A is not high curent ...
look at my addon board ..

It will be better for users .. have wire order one-to-one ...

EDITED:
it have 4 layers ..? (pcb)

High current in compare with current in analog circuits near.
Yes. Currently it is 4 layer.

Open air current hw/sw shows lambda of 2. All the way into lighter flame it shows lambda 0.6, with everything in between depending on flame position. Pictured lambda 1-ish

MCP6002 from JLCPCB definetly is not the best rail-to-rail opamp. Input is 0.00, output is 0.18.

Eh? We see much better performance than that on ECUs. Down in the single-digit millivolt range.

ps: nice HP scope

Yes. Initially I thought this is China voltage measurement unit issue. But not.

A good series of articles from @Maxi https://rotorman-dtt--motorsport-ru.translate.goog/j5-sport/innovate2.htm?_x_tr_sch=http&_x_tr_sl=ru&_x_tr_tl=en&_x_tr_hl=ru&_x_tr_pto=wapp .
Good description about pros and cons of "digital" sensor driving.