SaabCentral Forums banner

21 - 31 of 31 Posts

·
Registered
Joined
·
6,410 Posts
Mattis said:
My MB Quart RAA 2400 has input sensitivity 0.15-8V.

I should add that I'm not sure that AMP1 outputs 6V. I've been told 5-6V but it's unconfirmed.

/Mattias
Do you have this running yet? I assume if AMP1 is putting out ~6V you need to turn the gain on your MB Quart amp almost all the way down. Correct?
 

·
Registered
Joined
·
1 Posts
Obviously it works. Several users on the STCS forum in Sweden have their systems up and running. Make sure that your amplifiers RCA input (input sensitivity) can handle such high voltage as AMP1 is feeding. If it can handle 6V you are on the safe side I've been told.

The signal however is within the frequency range 20Hz-250Hz so it is NOT fullrange, but feeding a subwooferamp should work just fine.

/Mattias
Many thanks for putting up the links to those photos! looks like its gona solve my problem. I am in the middle of installing an alpine AMP in my 93 convertible which only has 4inch in the rear and looks like they have been programed to produce the midrange only. The amp I am putting is MRP-F600 which does take speaker level input, my question is that should I tap into the real speakers or should I follow the instructions in those photos to get the best sound! I am planning to put two 6x9 and a 10 inch sub.
 

·
Registered
Joined
·
282 Posts
Well, I did this - and I can't get it to work. The signal is very faint, and there's no bass.

I redid the job, and studied it - there's no reason it shouldn't work. I'm on the right pins, there are no shorts, and high level outputs still work.

WTF?
 

·
Registered
Joined
·
6,410 Posts

·
Registered
Joined
·
6,410 Posts
I would suggest giving up on the idea of soldering line-outs and just going with an LOC. That approach has a long track record of satisfaction here. The idea of soldering line-outs has a certain technical elegance I agree, but no one has ever shown that to be a superior way to get a usable signal, and we have so many happy customers using the simpleton approach. In short, why argue with success?
 

·
Registered
Joined
·
282 Posts
yeah - i have already abandoned that.

the only trouble now is the popping i get in the left channel!


I need a new AMP1, but I don't feel like dropping $300+ on it.

I hate this system
 

·
Registered
Joined
·
89 Posts
Well, I actually managed to get a properly working line-out on AMP1. It took some engineering but wasn't too difficult after all.



I have a somewhat detailed explanation on our Dutch forum. It is in Dutch however, which is probably not readable by all of you guys.

A brief summary of my "quest for a sub":
- I found that the PAL007 (a Pioneer proprietary power driver without available datasheet) is (at least pin)compatible with the TDA7560. However, you don't need to know too much about the pins since the TDA's input capacitors are marked on the bottom side of the pcb with EVOL, EVOR and EVOC (which is obviously left, right and center respectively).
- In a trial-and-error process, I tapped the input wires of the PAL007 but I found out that the amp I wanted to feed pulled the input level down too much, muting AMP1 (which is not what I want). So I added a simple line buffer (an opamp as voltage-follower). The schematics are on page 4 of said topic. To prevent interference with the driver, I took the signal before the input capacitor.
- This signal is just a perfect line signal to drive my amp, an old (1993?) Caliber CA-70, good for 70 clean watts which equivalents to about 150 car-manufacterer's watts. The amp now drives one TangBang 6x9" sub, which is sufficient for the time being.

Having the output up and running, I encountered two challenges.
- Firstly, the signal contained too much high-frequency, giving a rather confusing sound from the sub. To solve that, I added a 24 dB/octave Linkwitz-Riley filter just before the amp, summing the left and right signal (the amp is in bridge mode), which renders a sharp lowpass at about 126 Hz.
- Secondly, the subwoofer signal didn't mute as AMP1 muted (e.g. when having a phone call or when searching radio stations). I found out that AMP1 mutes by sending a 'high' (being about 4V) to pin 22 of the PAL007. On one of the pics you guys can see a place on the pcb to tap this signal. This was more difficult to tackle. I added a mute circuit to the buffer circuit in AMP1 consisting of a CMOS analog switch (CD4053) driven by a transistor, which for now works OK. There is still a slight audible noise ('rumble') on mute, but well below any level of annoyance. The final solution will be to use the mute signal to drive a small reed relay, which on its turn drives a bigger relay in the sub's speaker wiring. Of course this requires an extra wire to the trunk. Once I have the relay working, I'll probably remove the CMOS analog switch to eliminate the noise it induces.

I also measured some spectrums of the door- and rear speakers (picking up the front speakers was too much work). Apparently, there is big dip around 1kHz (of about 12 dB !!!), which accounts for the lack of 'presence' in the sound other 'premium' systems do have. I further found that the door speakers are cut off from about 120 Hz to -20 dB at 300Hz, which is a shame because these speakers with their size and box (the door) could make the system sound warm and rich at higher frequencies.

Hence my next project (if I ever have time for it) will be making a similar line-out on the EHU (at least the front L&R). Then I will sum this signal with the AMP1 signal I already have and feed it into AMP1's power driver (PAL 007) through a crossover (to be determined, but probably from 80-100Hz to 1200Hz) and perhaps a parametric EQ to lift the dip around 1kHz. Cutting off the low frequencies should be relatively easily accomplished by changing the input capacitors. According to the TDA7560 datasheet, the should be 100nF for a cut-off at 16Hz. Changing them into 22nF should (at least theoretically) move the cut-off frequency to about 72 Hz. Maybe I'll try that as an interim stage.

Yes, I'm crazy, but hey, the electronic components don't cost a lot and I find these projects very relaxing after my work in a office. Basically I agree with ctrlz that a LOC on the speaker output is the most practical solution for most people, avoiding the mute-problem.

Oh and because you americans are famous over here for your claim culture: this works for me, it mightn't for you. It requires at least basic electronics skills and knowlegde. If you want to try it, try it at your own risk and don't blame me if you fry your AMP1 or your entire car.;)
 

·
Registered
Joined
·
89 Posts
Line-out on EHU

Well, here you are! It was quite easy.

This is wat you get if you lift the cover:


Very friendly of Pioneer/Clarion/Saab or who ever designed the EHU to mark the power driver pins on the PCB:


No problem at all to find the right input pins. As with the AMP1, I took the signal just before the input capacitor to prevent pulling down the input signal too much:


It is quite a hassle to open the case, as one has to desolder and untwist five locks and five ground pins. This time I tried to get the +12V from the antenna amplifier wire. I hope it is silent enough otherwise I'll have to get through the worst part again:


The guys at Pioneer/Clarion/Saab must have known my plans, there's a nice hole in the PCB to lead the wires to the other side ;):


I ty-rapped the buffer PCB to the bottom side of the case where it sits against the CD-changer:


Drilling cinch-sockets doesn't make sense, so I made a cable ending near the 375 (carkit) connector:


That's it for now. I'm very curious what the output signal looks and sounds like. Still planning to sum this signal with AMP1's bass to get a full spectrum. I'll keep you posted!
 

·
Registered
Joined
·
89 Posts
I didn't try the sound yet, but I took some spectrums:

Full audio range (purple = white gaussian noise, flat spectrum; green is EHU's line-out; tone controls are flat), mind the 10 dB dip around 1kHz:


A closer look at 0-2,5 kHz:


And a close-up of the lower regions:


I'm quite convinced that the dip at 1kHz accounts for the 'cheap' sound of this system, whatever speaker upgrade you try,

Looks like EHU's signal is rather flat as of about 100Hz (apart from the dip), so I'll try to feed AMP1's power driver first with this signal and leave AMP1's source signal to the sub. Eventually I'll try to add a parametric EQ to fix the dip at 1kHz.

Next sub-project will be to make a line-in on AMP1 :lol:, but first I'll repeat measurement with a white noise mp3-file, to be sure the dip isn't caused by the Dension's line-in.
 
21 - 31 of 31 Posts
Top