When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
The subwoofer speaker wasn't working so I investigated
I found that one of the voice coils was cooked, charred, open circuit, and the coil former was seized up in the magnet
The speaker voice coil consists of a parallel winding (two in hand)
Each of the two voice coils is fed by a bridge amplifier, ie two bridge amplifiers into two windings on a dual voice coil
I rewound the voice coils but the coil former was slightly distorted from being cooked and not a perfect movement inside the speaker magnet
Even after the rewind the subwoofer speaker didn't seem to do much, so I have my doubts about the design of that speaker performing as the designer claimed
With poor audio output from the subwoofer speaker I then decided to check the subwoofer amplifier itself
When taking measurements I used my Tektronix 465 oscilloscope with differential inputs (so as not to earth one leg of the bridge amp)
I used my HP8920B radio test set and fed a signal into the car radio FM on a vacant channel at 88.3MHz to simulate the real world
Tests:
subwoofer amp - loads
3.9 ohm resistor loads at the plugs next to the subwoofer (two bridging amp circuits and two load resistors)
bridging amps - expected output
2x12v = max of 24v p-p across load for bridge amp
Tektronix 465 CRO differential inputs
HP 8920B FM 88.3 MHz into RX 75KHz dev
fairly loud from bass speakers in the doors
100Hz 5v p-p each speaker side of sub amp into load (0.8 watts)
very loud and booming from bass speakers in the doors
100Hz 10v p-p each speaker side of sub amp into load (3.2 watts)
So the sub amp seems that it can supply the subwoofer with adequate power
but the subwoofer itself is lame and the amp gain is too low
Conclusion
With ref to 30Hz
6dB cut-off around 90Hz
Half loudness around 100Hz
Roll-off 50Hz to 100Hz is 12dB per octave
15 watts into each voice coil
x2 = 30 watts into the double voice coil speaker
So sub amp power is good but the amp gain is too low and the subwoofer speaker is just a dummy load.
The speaker design is dubious with the cone facing down although it is said that it doesn't matter
There is also a large obstruction of air pressure from the suspension spider and plastic frame.
The subwoofer speaker is also just too small to push a large amount of air at those frequencies
This is just not a good design
The subwoofer also rolls off at say 12dB/octave
With that matched by an apparent gain of 12dB/octave gain from the sub amp filter the net gain is zero
If you've only got 0.8 watts at 100Hz that's all you're going to get in speaker SPL at 50Hz
So you need more sub amp gain to start with
If the subwoofer was replaced by one with better sound pressure output
and if the gain of the sub amp could be adjusted ie increased
then the subwoofer system may work well
Perhaps someone has delved into how to adjust the subwoofer amp gain (separate sub amp type)
Someone may have the specs for the amp ie the speaker impedance being used. The amp chips TDA7801 can drive 4 or 2 ohms
I found that the enamelled copper coil winding wire used for the voice coils was approximately 0.4mm dia and I rewound the coils with that
The TDA7801 power amplifiers state 28 watts 10% THD @14.4volts which ties in with my measurements of 15.5 watts at a lower battery voltage
0.4mm dia wire has a current rating of around 2 Amps, which with 4 ohms is 16 watts, so I'd say subwoofer A21282003502 is a 4 ohm speaker (two 4 ohm voice coils)
I don't think that the Harman 9010 subwoofer amplifier is capable of high power even with the cooling fan
to be continued...
Subwoofer after voice coil rewind (dust cover not fitted) part number A2128203502
Subwoofer amplifier Harman 9010
Last edited by Bruce Hubbard; 03-04-2024 at 08:48 PM.
So resigning myself to the fact that the subwoofer audio system installed in the car didn't work from the day it was installed
because the design engineers didn't apply enough gain to the sub amp or the drive to it is insufficient and the subwoofer itself is just a joke in design
I decided to try a 10 inch dual voice coil subwoofer from Fusion model number PF-SW100D4
Without going to all the trouble of installing it only to find it didn't meet expectations
I decided to test it free air and see how much cone movement we can get
Does it work?
Yes and no. It has more grunt than the piece of rubbish that came with the car but it's still not really thumping
and there is more bass coming from the front door speakers than there is from the subwoofer
There is no way of adjusting the subwoofer volume using the Comand screen that I can see
The sub amplifier chips TDA7801 seem to have gain adjustment from about 2dB to 16dB but under programmer's control
to be continued...
Last edited by Bruce Hubbard; 03-12-2024 at 02:32 AM.
From the Harman sub amplifier 9010 board layout and pinout for TDA7801 bridge amplifiers
we see they have used bridge amplifiers 1 and 3 and their digital signal in is on the I2S bus feeds SD1_3
the amplifiers have programming inputs on an I2C bus, they are not wired to the outside world, but are on the board not connected
the amplifiers are capable of being programmed for different gains (Table 6)
the code to program the gains is bits D3 and D4 (Table 11)
the programming sequence is shown (Figure 38)
But we do not have a programmer to read the data to know the amplifier gain
or to change the value of the amplifier gain
Harman engineers and product developers would have one
and also know the parameters, and we need to know what gain they programmed
to see if it can be increased or if the amplifiers are already running flat out gain
Will they share the information is the question
to be continued...
Last edited by Bruce Hubbard; 03-07-2024 at 09:30 PM.
whatever you can program the amp gain to, its gotta work with the supply power to the amplifier stage... odds are you can't get much more than it was preset to before you start to run out of pookah.
02-25-2024, 07:49 PM
