PicoScope 7 Automotive
Available for Windows, Mac, and Linux, the next evolution of our diagnostic scope software is now available.
Alternator ripple, what is normal?
Alternator ripple, what is normal?
Hi Pico support staff and fellow customers,
I hope I’ve presented my question in an appealing and understandable way.
So before posting I did a search for Alternator on this forum and found “Why all the noise!” from Steve Smith in 2018. Thanks for sharing that. I would like to follow up on that.
I’ve a Bosch 14V 55A alternator in my Peugeot 205 car. While installed in the car I measured the alternator voltage with my 200 MHz bandwidth scope. At that time I wasn’t familiar with the connections on the alternator itself so I measured at the battery.
alternator_spike2.png
One such a spike with a shorter time scale:
alternator_spike.png
So about 3Vpp at 10.5 MHz. This is worrying me. You know I’ve seen some YouTube videos about this topic and what they show is something like 200 mVpp, but maybe their scope bandwidth is much less than what I have.
Now the alternator is removed from the car so I was able to replace the two bearings. Totally unrelated to this voltage ripple issue. I made a bench test setup to be able to diagnose this ripple issue further.
I first show you setup A:
setup_A.png
This bypasses the voltage regulator and the rectifier so I can isolate the problem. I’ve soldered three wires (blue, green and orange) directly to the stator windings which where disconnected from the rectifier. The lab power supply provides a steady 1A field current to the rotor. I am measuring between two stator legs without any load with the differential probe at 20 MHz bandwidth.
1A_field_between_green_blue.png
So, the induced voltage from the rotor to the stator looks totally fine. No noise. I also measured between the other two legs, same waveform.
To avoid seeing the 20 MHz bandwidth as the cause of not seeing high frequency spikes I measured again on a regular 200 MHz scope input and providing 600 mA field current with a linear power supply. Still the same noise-free waveform even around 10 MHz.
Moving on to setup B:
setup_B.png
This includes the voltage regulator and the rectifier. Now the waveform at the stator legs looks like this:
stator_green_blue.png=
On channel A (blue) I have the alternator output voltage and channel B (red) I measure between the two stator legs. Again the two other stator legs display the same waveform. Anyway I do get about 14V charging voltage.
I measured this with a brand new rectifier the ARC0001 an aftermarket one from AS-PL for Bosch alternators.
Although I measure BEFORE the rectifier I do see significant effects of the rectifier. On the red signal I conclude I see diode reverse recovery spikes and the reverse breakdown voltage is 15V. But these effects do not seem to influence the blue signal, the output voltage. I suspect the random high frequency noise is caused by the diodes.
I did an AC-coupled measurement of the output voltage and waited for a big spike to sample:
alternator_rectified_voltage_11MHz_spikes,png=
So 10 Vpp at 11 MHz! Note the red signal is not the output voltage here.
I also did the same measurement with the original Bosch rectifier and played with different voltage regulators, the original Bosch as well as Hitachi and Metzger. In all cases big 10 to 11 MHz spikes.
So dear Pico support team do you have test setups for car alternators and if so what did you measure? Do my measurements seem correct and are spikes like this acceptable in the automotive industry?
Best regards,
Zeyneb
I hope I’ve presented my question in an appealing and understandable way.
So before posting I did a search for Alternator on this forum and found “Why all the noise!” from Steve Smith in 2018. Thanks for sharing that. I would like to follow up on that.
I’ve a Bosch 14V 55A alternator in my Peugeot 205 car. While installed in the car I measured the alternator voltage with my 200 MHz bandwidth scope. At that time I wasn’t familiar with the connections on the alternator itself so I measured at the battery.
One such a spike with a shorter time scale:
So about 3Vpp at 10.5 MHz. This is worrying me. You know I’ve seen some YouTube videos about this topic and what they show is something like 200 mVpp, but maybe their scope bandwidth is much less than what I have.
Now the alternator is removed from the car so I was able to replace the two bearings. Totally unrelated to this voltage ripple issue. I made a bench test setup to be able to diagnose this ripple issue further.
I first show you setup A:
This bypasses the voltage regulator and the rectifier so I can isolate the problem. I’ve soldered three wires (blue, green and orange) directly to the stator windings which where disconnected from the rectifier. The lab power supply provides a steady 1A field current to the rotor. I am measuring between two stator legs without any load with the differential probe at 20 MHz bandwidth.
So, the induced voltage from the rotor to the stator looks totally fine. No noise. I also measured between the other two legs, same waveform.
To avoid seeing the 20 MHz bandwidth as the cause of not seeing high frequency spikes I measured again on a regular 200 MHz scope input and providing 600 mA field current with a linear power supply. Still the same noise-free waveform even around 10 MHz.
Moving on to setup B:
This includes the voltage regulator and the rectifier. Now the waveform at the stator legs looks like this:
On channel A (blue) I have the alternator output voltage and channel B (red) I measure between the two stator legs. Again the two other stator legs display the same waveform. Anyway I do get about 14V charging voltage.
I measured this with a brand new rectifier the ARC0001 an aftermarket one from AS-PL for Bosch alternators.
Although I measure BEFORE the rectifier I do see significant effects of the rectifier. On the red signal I conclude I see diode reverse recovery spikes and the reverse breakdown voltage is 15V. But these effects do not seem to influence the blue signal, the output voltage. I suspect the random high frequency noise is caused by the diodes.
I did an AC-coupled measurement of the output voltage and waited for a big spike to sample:
So 10 Vpp at 11 MHz! Note the red signal is not the output voltage here.
I also did the same measurement with the original Bosch rectifier and played with different voltage regulators, the original Bosch as well as Hitachi and Metzger. In all cases big 10 to 11 MHz spikes.
So dear Pico support team do you have test setups for car alternators and if so what did you measure? Do my measurements seem correct and are spikes like this acceptable in the automotive industry?
Best regards,
Zeyneb
Re: Alternator ripple, what is normal?
You’re searching apples in a wrong tree. There’s nothing wrong with the car.
Voltage spike is induced by inductive loads switching off. Bring energy in the system and voltage goes up if not dissipated. And that’s not from the alternator. !!!!!!!!!!!
Voltage spike is induced by inductive loads switching off. Bring energy in the system and voltage goes up if not dissipated. And that’s not from the alternator. !!!!!!!!!!!
Re: Alternator ripple, what is normal?
I would agree with that.You’re searching apples in a wrong tree. There’s nothing wrong with the car.
Voltage spike is induced by inductive loads switching off. Bring energy in the system and voltage goes up if not dissipated. And that’s not from the alternator. !!!!!!!!!!!
Would ge good to see a PSdata file of Alternator ripple, Injector 1 Current & Coil 1 Current (if petrol?
Re: Alternator ripple, what is normal?
I would agree with that.You’re searching apples in a wrong tree. There’s nothing wrong with the car.
Voltage spike is induced by inductive loads switching off. Bring energy in the system and voltage goes up if not dissipated. And that’s not from the alternator. !!!!!!!!!!!
Would ge good to see a PSdata file of Alternator ripple, Injector 1 Current & Coil 1 Current (if petrol?