No READY. P1A15 error. Condenser charge timeout.

Mitsubishi i-MiEV Forum

Help Support Mitsubishi i-MiEV Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
kiev said:
Was your op amp definitely bad--let us know how this works out for you.
yes, after replacement AD8677 the car started up
but not for long.
As I understand it, the problem is in the capacitors, which, after heating, give an error again
 
kiev said:
i have had problems with intermittent ceramic capacitors due to temperature changes. Hard to troubleshoot too.

replaced r211,212,216,218,219,220 + c227,229,231
left to be replaced r221
problem still exists
 
Here's some ideas. With a voltmeter carefully measure the voltage of pin 5 of the hybrid board with respect to pin 4. i would expect it to be equal to about the aux battery voltage.

Would you have access to an isolated oscilloscope (not a common item) to be able to monitor the Input (VH with respect to VL) and the Output (pin 3 with respect to pin 4) signals of the Hybrid board? An isolated scope would prevent shorting the chassis ground to HV negative while allowing both signals to be captured during the start transient.

If no isolated scope, then two separate captures could be done, one with only the VH to VL signal, and another with only the Output (pin 3 to 4). This would point to an issue with the Hybrid board if the Input goes to an expected level but the Output does not. Be careful with touching the scope ground lead to anything but the desired signal return; maybe solder some extension wires from the desired terminal pins to allow better access and also use electrical tape to control accidental touching/shorting.

If no scope at all, then possibly the values could be observed on a voltmeter, but the transient is likely too quick to really get a reading. However it would be worthwhile to measure the Output with a meter to see if the signal is pegged high or low with the key at ON position and then at START. This could indicate an issue with the final stage 4066 ic or the Low Pass Filter.

From Ged's thread, bottom side of hybrid board with markup:
aydmOQY.png
 
i measured the C227, C229 and C231 surface mount capacitors as 390 nF (100Hz) for a combined ~1.2 uF at the pin 1 input of the ISA215 board. They are in parallel with the little blue precision voltage divider resistor R221, which was 6.8k in the MCU that i have dismantled. So the rise time constant is ~0.72 sec and 3x is a little over 2 seconds. So i'm guessing that the ECU waits at least 2 seconds after precharge to sense the Vcap.

i suspect that one of those capacitors is leaking. It can be tested on the bench with a 30VDC power supply at the via by R220 and measure the voltage divider at pin 1 of the ISA board. then heat those capacitors and watch the voltage for drop.][/quote]




Can you help me with the test for the capacitors please

where do i connect 30v dc to exactly and is the voltage divider the chip on the hybrid board with white markings?

thanks again
 
~2 seconds is a good value; the best example of the fault is found on page 11 of this thread in the scope capture by czerodk, http://myimiev.com/forum/viewtopic.php?f=23&t=4225&start=110#p42510

Notice that the voltage doesn't reach high enough and there is a a little delay step at 300V on the way up. That little notch is causing the fault. And all these threads are searching for ways to find what is causing that step/delay/notch thing.

My thoughts on the "30V" test were to apply a voltage toward the bottom of the resistor divider (blue resistors on the control board) that would simulate having high voltage at the top end. The input voltage to the hybrid board should never go over 5V and in normal operation will be closer to 4V. The 4V signal can be monitored while heating the 3 capacitors with hot air to see if it shows any weirdness. The via near R220 seemed like a good place to solder a small wire to apply a test voltage; another approach would be to solder near the 6.8k resistor and just apply 4V there.

If the capacitors can be removed and tested on an LCR meter and heated during those measurements, then that would do the same thing and be a better approach; but not many people have access to an LCR.

But the easiest thing may be to just replace the caps, all the caps involved in this circuit both on the control board and on the hybrid board, which requires desoldering the hybrid board from the control board. And also replace all the ic chips on the hybrid board.

It is very difficult to get an intermittent fault to occur when you want it to, and a simulated high voltage may not excite the fault, so much time could be spent chasing phantom blips. Also it could be something else besides the caps, such as the op amp getting flaky under thermal changes.

Nobody has isolated the root cause; i don't have a bad board to do the testing on myself, all my boards work, so i'm only able to make suggestions of things that i might try to do. i have found and repaired ceramic capacitor failures in car electronics, and these devices have a bad reputation for intermittent failures, just do a google search to see.
 
All connected ready on a test bench

What am I expecting to see at pin 3 and what am I measuring? Volts or a waveform

Also where else do I put my test meter?

Lastly how do I post pictures!?!


 
Images have to be hosted somewhere and then linked to from the forum post; lots of us use imgur.com for the hosting service.

In this case a picture would be helpful because i have no idea what is your configuration for testing--are you thermal testing the 3 capacitors on the control board or have you removed the hybrid and testing it separately?

Would it be correct for me to assume that you are an electrical engineer with all the knowledge, skills and abilities to troubleshoot electronics, read schematics, device part numbers and datasheets, and trace circuits on a pcb, with a full complement of power supplies, meters and scopes available for measurement and testing?

The chip with the white lettering on the hybrid is the Op Amp, the lettering is a date code.
 
Hi

I wanted to test the hybrid board still on the main board as it left the factory without disturbing anything first.

What I want to do is pinpoint the fault before attempting a repair, I am concerned as soon as I de-solder the hybrid board I could make the problem go away for a bit also I would rather test as much of the circuit as possible prior to the hybrid board,

I have done all of the specified testing as previously mentioned and as per the schematic diagram, I do have all of the equipment and quite an experience of in depth electronic diagnosis and chip cloning / programming.

I have set this up on the bench with a variable supply voltage to VH / VL and a 12v supply to 4 and 5.

I am now going to try and post some pictures!
 
Sorry for the barrage of posts!

The output voltage on pin 3 is 0.15v lower than the input voltage on VH / VL. Is this meant to be exactly the same?

I want to check this before I warm or cool any components

I feel like I am getting somewhere and if the voltages are meant to be the same then at least its easier to make a diagnosis on the bench.........
 
That's an excellent test setup and now i understand your approach. You are exactly set up to find this fault.

The output voltage on pin 3 should be exactly the same as the input voltage on pin 1. The fact that pin 3 is lower is the cause of the P1A15 DTC. What we don't know is why the signal is degraded or getting attenuated by the hybrid circuitry. Is it a bias due to a defective part? Does it vary with temperature? ...What sort of part degrades without failing?

There are several ceramic chip capacitors on the hybrid, on the upper and bottom layers. Maybe the upper side caps, resistors, chips could be thermally tested while monitoring pin 3.

How much current does the 12V supply pull? After an inrush spike my board settles out to 20mA.
 
Thanks for the quick reply

To start with is there any point taking the board off, I would rather leave it on for as long as possible without aggravating with heat.

I have not applied any heat to it as I have never seen this car run and the temperature is a constant 20 at the moment so is there any point applying heat or cold? What I dont want is the fault going away!

Can I take any more measurements from anywhere else on the board or shall I now remove the capacitors, test and replace?
 
Sorry I forgot to put that I dont know the 12v current consumption as I only have one variable power supply, the 12v supply I am having to use a fixed voltage supply with no amperage measurement. I can connect an inline ammeter if this is needed
 
No just curious if the 12V was getting pulled down then it might affect the board operation.

You could measure the output at pin 6 of the op amp, it should be inverted equal to the input voltage signal to the hybrid pin 1.

also the low voltage supplies created from the 12V. These are at vias coming off of the 5-pin chip, pin 3 is the V- and pin 5 is the V+, created from a rectifier on the bottom side. Pin 2 is the excitation and should be an AC waveform, pin 4 is an inverted output (AC or 0 to 5 step or ?) signal that feeds the control gate on the bottom.

Other interesting measurements would be at the two vias above the chip on the output side of the hybrid.
 
Hi all,

Can theoretically the signal transformers on the hybrid circuit be demagnetized so the gain is reduced and then the voltage signal too?

We should consider that these transformers are close to the motor that should generate a strong magnetic field and after so many years some effect could happened.
 
in this picture pin 1 is the lower left pin of the chip with "B11", it is an AD8677 op amp. pin 6 is on the upper row above the last "1" of "B11"

UFACkBPl.jpg


The two vias above the edge of the chip on the far right. i think the left via will be a reference voltage level determined by the two resistors to the left, and the right via is 12V.
 
Back
Top