Leaf - Gen 2 PDM/OBC/charger no output power - L1/L2 charging

UCC28070

• Pin 2 - Rrdm = 137 kOhms
• Pin 4 - Vsense = TP2164 = 13 kOhms to GND (DC-) = Rb
• Pin 4 - Vsense = TP2164 = 3.42 MOhms to Drains of Q100, Q101 = Ra
• Pin 4 - Vsense = TP2164 = 1.95 MOhms to Drains of Q102,Q103 = Ra
• Pin 5 - Vinac = TP2165 = 1.579 MOhms to Vin = Ra - it's weird these values are different even though the Typical application doc says this should be the same as Ra above.
• Pin 5 - Vinac = TP2165 = 13 kOhms to GND = Rb
• Pin 6 = Rimo = 55.5 kOhms to GND
• Pin 7 = Rsynth = 80 kOhms to GND
• Pin 8 = Csb = 491 ohms to GND
• Pin 9 - Csa = 491 ohms to GND
• Pin 10 = Pklmt = Rpk2 = 9.16 kOhms, meaning should be Rpk1 = 2.84kOhms, but Rpk1 = 6.92k and Rpk2 = 9.2k
• Pin 13 - Vref = TP2149 = 12 kOhms to GND = Rpk1 + Rpk2 = 2.84kOhms + 9.16kOhms
• Pin 14 - Gda = TP2147 = 21.2 kOhms to Gate of Q102, Q103, Q100 and Q101, Pin 14 to Pin 17 is 42 kOhms, through R174 (509 ohms) to "In A" of MCP1404E
• Pin 15 - VCC - goes to Pin 7 of PFC connector (CN2001)
• Pin 16 - GND - (Pin 17,15, 9, 10, 8, DC-)
• Pin 17 - Gdb = TP2148 = 21.2 kOhms to Gate of Q102, Q103, Q100 and Q101, Pin 14 to Pin 17 is 42 kOhms, through R176 (509 ohms) to "In B" of MCP1404E
• Pin 18 - SS = TP2146 = 838 kOhms to GND. C122 to GND Doesn't got to PFC connector to be used as an external fault stop.
• Pin 19 - Rt - sets the switching frequency = 90kOhms to GND
• Pin 20 - Dmax = Rdmax = 95.3 kOhms to GND

Now,
To test input Voltages into these ICs, which can be done safely now that I know what Pins they map to on the PFC connector CN2001.

kennybobby said:

Is the circuit board discolored or scorched in the vicinity of some of the FETs, or is that shadows?

Click to expand...

No, it's just around the soldered joints and can be scraped off. No damage to components.

kennybobby said:

There is a resistor in parallel that shunts the diode check measurement.

Click to expand...

Great that makes way more sense.

Drew up the H-bridge (inverter) circuit. Looking to replace C507 and C509.

C507 is the same as C505
C509 is the same as C510

- Found C509 to be shorted at 12ohms.

IRS2113S - The IRS2110/IRS2113 are high voltage, high speed power MOSFET and IGBT drivers with independent high-side and low-side referenced output channels

• Pin 1 - Lo - Gate of Q503 and each IC respectively to Gate of Q501, Q2003, Q2001
• Pin 2 - COM = GND for all = Pin 5,6,7,8,11,12 of CN2000
• Pin 3 - Vcc = Vdd = Pin 9 and 10 of CN2000, also equals Pin 11 of CN2001 = Vin of MCP1404E
• Pin 6 - Vs = Source of Q2000 and each IC respectively to Source of Q2002, Q502, Q500
• Pin 7 - Vb = Diode checks good to Vcc (+)
• Pin 8 - Ho = Gate of Q2000 and each IC respectively to Gate of Q2002, Q500, Q502
• Pin 11 - Vdd = Pin 9 and 10 of CN2000 - goes to all IC2000, IC2001, IC503, IC504
• Pin 12 - Hin = 21.5 kOhms to GND
• Pin 13 - Sd = 0 ohms to GND
• Pin 14 - Lin = 21.4 kOhms to GND.
• Pin 15 - Vss = GND (Rectified DC-)

Traced the Vin for the UCC28070 chip that wasn't getting power during my tests. This is critical for the PFC to work properly.

I'm looking a datasheet on the component marked with "4M". I has 4 terminals. 3 on one-side and 1 on the other. the middle terminal has continuity to the lonely pin on the otherside. I assume it's a transistor. 1 = Base, 2 = Drain and 3 = Source. Pin 1 of "4M" has 5.8kOhms resistance from Vout of the 13.0volt LDO, A1J130 markings with part number: R1501J130.

** Which this ground is Pin 2,4 of CN9006 which is also the same as (Pin 17,15, 9, 10, 8, DC-) of CNCN6902/CN2001
and for UCC28070 - Pin 16 is its Ground and goes to (Pin 17,15, 9, 10, 8, DC-) CN2001.

kennybobby said:

Is there any resistance to ground from TP9037 between the 4.7k and 1k resistors on the "4M" chip? If 4M is a transistor then there needs to be some voltage divider path to supply current to the base.

Click to expand...

Yes, 11k to GND. See the control circuit below.

This control circuit goes through a series of transistors and ultimately into Pin 5 of CN9006. Pin 5 is tested to have 5V when OBC should be be charging. If Pin 5 is high, it should make T4 closed, T3 closed, T2 open, and therefore supplying +13V through a R1 = 5.7k and R2 = 11.7k voltage divider to get 8.7V at the Gate/Base of T1 to drive T1 closed that sends 13V to Pin 7 for the Vin of UCC28070. What's actually happening, is T2 is closed, and TP9037 is grounded through 188ohms meaning gate.base of T1 has not enough voltage to drive it closed.

• Pin 5 goes into Pin 14 (OUTa) of IC351- 3AATQQT or 3AAT00T 1S07241CQ - HIGH-SPEED QUAD DIGITAL ISOLATORS
• Max voltage is 6V and we're getting 5V, so i think this part is doing it's job. (I thought this would need to be higher to drive T3 closed)
• Now, need to figure out why T2 is closed during charging. My best guess right now is because there isn't enough voltage/current at the gate of T3 to close it.

kennybobby said:

Where are the T1-T4 located--is there a picture posted of that board?

Click to expand...

See below:

Another couple measurements i was able to get were:

• T2 drain = 0.786V when charging
• T3 drain was 0V while charging
• T4 drain was 0V when charging
• Traced another circuit to T3 gate. Goes through 8.32k and goes to Pin 8, up to the top-side of the charger, into Pin 2 of CN5102, then onto the L1/L2 input charger board after the black relay. TP2000. If this circuit gets 12-13V then T3 would close and therefore leave T2 open. And since the L1/L2 input board only has Low-voltage one connector this must control the relay. This doesn't make sense because that relay is definitely closed because we're getting rectified DC at the output of the L1/L2 input board. Going to investigate this.

On the early gen units, the PFC chip senses if the AC is present thru the pre-charge thermal resistors. If so, then a signal is sent to the microcontroller that engages the drive circuit for the black relay coil. If not, then it sends a "cancel charge" back thru the system to shut off the AC provided from the EVSE.

i haven't measured it on a scope, but i suspect this AC determination is made within 10-20 msec such that the thermal resistors will not be overwhelmed. Quite a few early gen versions had a failure mode of blowing the fuse in one of the thermal resistors or of a resistor cracking open, when the AC relay was not engaged but the AC lines were still "hot" from the EVSE. Under normal operation those resistors should not carry very many cycles of AC.

On the TCCH chargers this is all done analog using a ViPer chip that bootstraps up the low voltage supplies to engage the relay coil. If the Viper fails, then the precharge resistors also will blow and make a loud pop.


Where does pin 1 of CN5102 connect--is it just tied to a local return/common ground plane?

Those little electrolytic cans on the control board may help give you some clues about the different low voltage supplies created on that board from the 12VDC supplied on the input connector, likely one or more 5V supplies, 15 to 16V supply. measure all those cap voltages if you can. There may be more than one 12V feeds, one is hot all the time and another is switched by the EV-ECU when charging is intented to start, and maybe another for when the DCDC is intended to start.

kennybobby said:

Where does pin 1 of CN5102 connect--is it just tied to a local return/common ground plane?

Click to expand...

Correction: CN5401 Pin 8 is 100ohms to GND. Pin 6 is +15VDC and should be available as soon as it's available. This drives RLY 1 closed through the 177ohm coil. Traced this circuit below.

- I also jsut traced T3 gate through a 14.6k ohm resistor to +13V LDO output. Voltage divides to 4.8V input to T3 gate So that means the UCC28070 chip should get power when +15V is supplied. So +15V must be switched somewhere. Which is happenning correctly. So im not chasing this +15V

- T3 drain is 0.786V, which should be enough to keep T2's gate open. And therefore driving current through T1's gate. But at T1's gate i'm getting 0.786V as well. Where the circuit should read about 9.48V at TP9037 and 8.7V at TP9038 aka T1 gate.

kennybobby said:

On the early gen units, the PFC chip senses if the AC is present thru the pre-charge thermal resistors.

Click to expand...

Are you talking about the Vsense pin on the UCC28070? If so, I would think that Vcc should always be powered and it would soft-start whenever there is AC sensed through the bypass precharge diode?



I think the "4M" transistor T1 is 2SC6126

Good catch on the 4M

The early (2011-12) gen1 versions of the OBC used a different PFC chip than in your OBC.

Seems like someone figured out how to run the PDM with gen 2 OBC on the bench, but i don't recall the thread or details. May be a combination of discrete 12V lines and/or CAN buss commands.


Where is that 100 Ohm resistor located that is connected to the relay?

Do signals on connector CN5401 exit the PDM or is it entirely internal?

Is IC5401 associated with the relay drive voltage? What is the Vin and Vout of IC5401

i noticed that the control board has isolation bands in the circuit planes.

kennybobby said:

Seems like someone figured out how to run the PDM with gen 2 OBC on the bench, but i don't recall the thread or details. May be a combination of discrete 12V lines and/or CAN buss commands.

Click to expand...

Here's Damien's video:

As well as 49thdriver on mynissanleaf.com:

• Stand alone OBC/PDModule EV system Can 2015 - SOLVED - My Nissan Leaf Forum
• EV-Projects/LEAF_OBC_StandAlone_v3.ino at main · 49thdiver/EV-Projects
• I also re-created that Arduino C script for the RPI, started charging normally (so the HV contactors closed) then injected the script (which is sending specific CAN BUS commands and specific intervals) into the can system but never got any current draw.

kennybobby said:

Where is that 100 Ohm resistor located that is connected to the relay?

Click to expand...

The 4x 391 ohms resistors in parrallel on the bottom rleft of picture.

kennybobby said:

Do signals on connector CN5401 exit the PDM or is it entirely internal?

Click to expand...

Not directly out of the PDM through CN100. (PDM connector F23 as well). and all power supplies coming through CN5401 are accounted for to get the PFC IC's fully powered. Pin 1,3 = 15V.

kennybobby said:

Is IC5401 associated with the relay drive voltage? What is the Vin and Vout of IC5401

Click to expand...

No, both those identical ICs power some smaller ICs on the top of the board. Pin 5 powers the IC right next to CN6902/CN2001 (going to PFC, very top right of last picture). I'm not sure what Pin 13 powers. Vin is 5V and each I/O channel is 5V and haven't traced them anywhere. Probably back to the big controller chip.


As of right now, the only irregularity I have is that the PFC controller UCC28070 is not getting Vin. This comes from Pin 5 of CN2001/CN6902 and goes through a series of transistors that should be switching to get ~13V to Vin.

I did some math on T1, and T1 seems to be operating in Common Collector configuration, and Vemitter/Vin for UCC128070 (13V from LDO) should be about ~8.04V.
Where, Vemitter = deduced voltage divider - 0.7V = 13V*(11.7k / ( 5.7k + 11.7k)) - 0.7V = ~8.04V.
- But i'm seeing was 0.789V at Vin into the Base of T1, and therefore should see 0.789V-0.7V = ~0.089V at the Vemitter. aka Pin 7 going to Vin of UCC28070.

Knowing that I might be right on this math, i'm going to go measure some more live charging values. Specifically:

• Resistance to GND of TCP9037. aka Collector of T2
• What is the Ve of T1. I know Vc is +13V and have measured Vb to 0.789V in past.
• Voltage at T3 collector (top terminal)
• Resistance to GND of T3 collector (top terminal)
• Live charging means, the entire system setup without CN6902/CN2001 connected, and rectified 330VDC at the PFC input.