PPI PC2350 RF Noise or Self oscillation problem
I have a strange problem with PPI PC2350 car amplifier.
This amp is not a new one, was repaired by me some time ago (changed some power transistors), but also repaired by someone before me.
The problem is this:
When I power up the amplifier (by power supply and remote pin), all radio stations, received by car radio loosed quality – brings some noise as background, some stations disappears at all.
Checked this with different radio receivers – result is the same.
I measured the RF field around the amplifier with spectrum analyzer – there are something like white noise around the whole frequency range from 50MHz to above 1GHz with no spikes on specific frequencies – almost constant distribution of noise level around the whole frequency range of used equipment.
At first sight it looks like self oscillation somewhere outside measured frequency range, or high electromagnetic field, that makes my spectrum analyzer to go crazy
I measured this noise/field from other car amplifier (Diamond Audio model D7152) – there are no noise, no field, spectrum analyzer shows no difference when amp is on or off, and radio works perfect.
This means I have a real problem with this amplifier.
The amplifier’s bias current was set to 0mA – Class B before me (from some other guy).
I changed it to about 10mA per transistor (1.5 mV across the emitter resistors), and now amp works in Class AB, with less distortion, but the RF noise is the same.
Idle current with no signal is about 3A with 12V from accu.
After DC-DC converter I measured +/-56V… (It’s different from values from some PC2350’s schematic that I have. There is written +/- 50V. If someone have an idea witch one is correct ?!? I can try to tune to 50v even if there are no trimmer pot. for this voltage…)
From the output, I can’t see some oscillations or big noise… looks fine for car amp. There are some HF spikes with level of about 150-250mV (f=40, 80, 100KHz) , comes probably from power supply.
The output without signal and load looks like this:
I checked with oscilloscope some points shown here:
AC voltage on positive and negative rails to GND2 (measured +/-56V (+/-50 on schematic ???))
Voltages on the both inputs of the inductor L1 to GND2 (after rectifying diodes – pin 2 and pin 3 on L1):
Voltages on the outputs of PSU MOSFETs to GND1 (common S-D of Q11,12, Q5,6, Q67,68, Q73,74)
(GND1 & GND2 are grounds before and after pulse transformer T1.
Is the heatsink grounding screw in place and tight?
Is the antenna shield ground good?
If this is the screw, that is near to the 12V DC input (with rectangular solder pad under the screw - connected to the ground (-12v)) - yes. It is tight - I checked right now - resistance between heatsink and -12v is less than 0.3 ohm. (may be less, but may ohmmeter's lowest range is 200 ohm ).
The antenna is the original HONDA active antenna, that uses rear window heater as antenna, and small amplifier near it. (btw, we changed this antenna amplifier before we found that the problem comes with PC2350, but the same result.)
Now, I'm checking also with standard home radio reciever (tuner) with external antenna - the same results. 2-3 meters around the amplifier - no good radio stations reception...
I checked this with other car amplifier, with many other (PC and other) switching PSU's around, but no problems with all of them.
I don't see anything out of the ordinary on the waveforms you posted. Maybe 1moreamp can offer some suggestions.
Have you tried the amp in another vehicle to see if it causes problems?
Yeah Perry, I have seen similar issues in the past and it was the power supply transmitting.
First things first is to check the zobel networks on the switching mosfets R81 & C52, R80 & C50, R183 & C126, and R196 & C148< the RC networks across the mosfets they are easy to find they are all 2 watt sized resistors> They are supposed to cancel out switching noise. The 2350 being a full H bridge power supply makes it one of a kind out there in the car audio world.
Also I would lift the board out of the case and inspect the entire circuit board for broken and burnt traces, especially ones that connect to ground in any way.
I am very doubtful that there any issues with the audio section, since this problem is born of the power supply's failure to filter out switching noise common to switchmode power supplies.
If none of that provides to be fruitful then we may need to look all the electrolytic caps also. You will notice all of the 2200 ufd caps also have 0.1 caps across them. I would check each of them also.
PPI used ferrite beads on all of there diodes to isolate switching noise at the diodes, and it worked as I recall. Common RF blocking is to add ferrite beads to the leads of mosfets, and diodes. Seen that for more then twenty years now. And lastly if none of this has helped I would start looking at the HIP-4081 mosfets drivers perhaps one of then is contributing to the issue.
I will be offline till tomorrow sometime, so please bear with me if I am not responding till then. this list above should keep you busy till then :)
Thank You Perry Babin & 1moreamp.
Tomorrow I'll check all this things.
All of them are possible reasons for this problem.
I'm already checked zobel networks resistors, but only from the top part of the pcb. It's possible to have burnt traces on the bottom side around power resistors or elsewhere... but let me check tomorrow when I'm back on work.
If you have a copy of the PPI PC2350 schematic produced by PPI, I'd be willing to buy a copy of it.
BUMP any new info on this thread ?
The same situation with my amp(pc2350).
At the moment both amps are together on the soft4gsm working desk :)
So... I'm a bit disappointed :(
I measured / tested Staykov's amplifier, and the situation is almost the same... The same noise on spectrum analyzer, the same problem with radio reception when antenna is near the amp... little less spikes / switching noise (resonance noise) around edges of voltage/current of switching transformer/mosfets, but the result is the same.
There are small difference in transformers in both amps - one is with multipair (2 wire) secondary winding - the second is with only single wire.
Also there are different ferrite beads on rect. diodes - on one amp they are 2 times longer than on the other amp.
This is for now. I'll continue on Monday next week.
|All times are GMT. The time now is 07:39 AM.|
Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 17.65%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio