Alpine MRV-F307 Tweaks

[RyanW]

My older Alpine MRV-F307 4-channel amp is in need of repair... While it's torn apart, I might as well tweak it a bit! I am currently using the line level inputs on the amp as I don't want to change my Headunit from stock (It just invites thieves). My plans are:

1) Use an LOC rather than the amps speaker level inputs. I'm thinking of building my own so that I can control the quality of components. I've thought of tapping into the Headunit ahead of the amplifier stage, but I like the idea of a balanced output because interference is a big problem (Cell phones, etc...). Why is the impedance of an LOC so low (~20 ohms), can I go higher (~100ohms) to ensure that the headunit amp operates in Class A?

2) Replace any signal path electros with MKP's

3) Add some capacitance to the rails. Does this work with an SMPS supply? Can I add some 1uF low ESR snubber caps or will this promote oscillation of the supply?

4) Increase the bias

5) Use good shielded cable from HU to amp. What is recommended? I have been thinking of the following:
- Cat 5e (One channel per twisted pair) Cross talk?
- Cat 5e (One channel per cable, solid=+ve, striped=-ve) Capacitive load?
- Balanced XLR style wire (Expensive)

But, before I can do anything, I would like to find a schematic... I've searched all over the place but can't find anything. Does anyone know where I can find a schematic for this amp?

Thanks,

Ryan

 

[1moreamp]

Pacific Coast Parts in SoCal. They are a Alpine parts distributor I am sure they can sell you a copy if no one has one here. Just Goggle the name to contact them. There website is passworded but you should be able to get a phone number...

 

[RyanW]

Thanks for the info. I checked there website... $10 for the "service manual". If no luck here, I will buck up and purchase one.

 

[Perry Babin]

The pacparts site isn't password protected. To get past the login screen, click 'submit' without entering anything in the fields.

http://www.pacparts.com/

The amp mods you mentioned aren't likely to make any audible difference unless the mod replaces a defective component. If the amp is in proper working order, there is very little you can do to improve the sound quality.

All stages of the head unit are likely to be class AB and there is nothing you can do to convert them to class A. If you want balanced output, use a ground loop isolator instead of an LOC (or open the LOC and remove the loading resistor).

The inputs of the alpine amp aren't likely to be balanced so having a balanced signal won't do much. The transformers in the LOC or ground loop isolator could alter the sound. You may want to tap off of the signal in the head unit and ground the shields of the output cables to the case of the head unit. This is how most aftermarket decks ground their shields. The problem will be to get enough signal. The signal levels in the head unit may be too low to use.

 

[RyanW]

Thanks Perry,

I agree that the headunit is Class A/B, but correct me if I'm wrong; When driving a high impedance load, if the bias current is not exceeeded then the amp will operate in push-pull class A. Thus, if the bias current is 100mA and if the headunit is only pushing ~8V (max) across 100 ohms = 80mA then there should be no cross-over distortion because the transistor never switches off.

Assuming that the headunit output is bridged, which it likely is, then the headunit speaker level outputs can be treated as balanced outputs. Feeding these balanced outputs into a transformer should act similarly to feeding it into a differential amplifier (ie. any interference applied to both signals will hit the transformer simultaneously but the only part of the signal that the transfomer will pass is the portion differing between the two... similar to a differential amp which is the backbone of a balanced input)

I don't really know the application of this stuff; I am just trying to apply the theory to the best of my understanding.

 

[RyanW]

The part I am trying to understand is: Why don't all LOC's use a higher impedance ~100 ohms? Every converter is 15-20 ohms that I have seen... There must be a reason for it. The 10W 20R is going to be more expensive than the 1W 100R so there must be a reason that lower resistance is used... My guess is that it has something to do with amplifier stability. The lower resistance will be more effective at damping any oscillations. Is this correct?

 

[Luke]

whats a LOC?

 

[1moreamp]

Well we use a Password to get in to do our orders from them, So I know of no other way to enter their website.. We have regular customer status account....

 

[Perry Babin]

RyanW:
Crossover distortion won't be a problem. As long as there's sufficient idle current to keep both transistors on at very low signal levels, there's no way for there to be crossover distortion. If there's no audible distortion at 4 ohms, there certainly won't be any at 20 ohms.

I believe the resistors are there for amplifier stability. Most amps will probably be stable without it. a 100 ohm resistor should provide the same protection against oscillation for most amps. They probably use the 20 ohm because it's virtually 100% guaranteed to prevent problems.

In a balanced system, the line driver AND the receiver have balanced connections. This is important for the balanced system to be of any benefit. Your amp probably doesn't have balanced inputs. If the transformer was at the rear of the vehicle, it may provide the same benefits as a balanced system. Near the head unit, it probably won't. The problem with having the converter in the rear is possible noise issues. Sometimes, you have to ground the brown wires of the LOC to the head unit's chassis to prevent noise. Grounding them in the rear of the vehicle also works sometimes but not always.

If the converter works (with no engine noise) in the rear of the vehicle, you could probably use virtually any type of wire between the head unit and the LOC. Tapping off of the speaker wires in the rear of the vehicle should be fine.

In your case, you need the transformer to convert the DC biased signal to an unbalanced signal with no DC. The possible benefits of a balanced system/noise cancelling function are secondary.


Luke:
Line Output Converter It converts a speaker level output to a preamp level output.


1moreamp:
Most people here won't have an account and I didn't want them to think that they would be locked out of the site.

 

[jol50]

That is the problem with factory HU, most don't have the quality. You put junk into an amp and it just makes louder junk. Maybe newer are better, I have not done that in a long time. Now most are processed too, like the Fords that lose all the bass above half way....everything is foolproof junk. They even cut the motor at 4K in park.

 

[tsmith1315]

Perry:

They probably use the 20 ohm because it's virtually 100% guaranteed to prevent problems.

Along a similar line of thought, it's closer to the load the HU's internal amplifier is designed to drive. I would expect better all around performance (SQ), even if marginally so.

Ryan:
My experience with the older Alpine speaker inputs was that they worked every time and sounded surprisingly good. The only others that consistently worked as well for me were on the ADS 642 crossover.
You could certainly DIY a better quality component, but it's questionable whether you'd hear the difference when the source is an IC amp in the HU.

As Perry suggested, tapping off the HU would yield a better quality signal. I've done it a few times in the 500mV days with good results.

What kind of factory unit is it? Perhaps someone here has had experience.

As far as cat5 goes, Adire did a little research into that and found medium quality coax delivered something like 10dB better performance against inductive coupling and 30dB better against radiated (engine) noise. I've got the .pdf somewhere if you're interested.

 

[RyanW]

Thanks for the feedback guys; I am interested in any documentation that will help me learn what is going on. I currently have the 4 channel amp under the driver seat with the HU connected to a star ground at the same point. This is where I would connect the transformers/LOC. My car is a 2005 Mazda 3... Any feedback on the Headunit is much appreciated. I like having steering wheel controls and factory appearance to prevent thievery.

I am fairly cetain that the problem I am having with my current setup can be attributed to the fact that I am using the low quality speaker level inputs onboard the amp. When I pull thngs apart, I will either upgrade or eliminate this portion of the amp.

If I were to tap off the HU, would you recommend a small opamp circuit? Perhaps a pair of OPA 2132's would suffice?

 

[Perry Babin]

Visit a couple of the various mazda forums. Someone should be able to tell you what you need to know about the head unit's sound quality.

The speaker level inputs on some amps are very low quality. Generally, the problem is engine noise. In the past, I've found that the LOCs generally worked better.

If you go into the head unit and cannot find a point in the circuit that has sufficient signal, you would need to use an op-amp. You don't need to use expensive op-amps. A TL072 will do the job. If you mount the op-amp in a socket, you can experiment with various ICs.

 

[I Am An Idiot]

You have to be old to know why the 20 ohm resistor is better than the 100 ohm in a Line Out Converter. If you had a radio with a 2 channel output chip and a resistor/heatsink fader, and you connected 2 LOCs with a 100 ohm resistor the fader would not function properly. You would always have output from both converters. With the 20 ohm resistor when faded to front the rear would have no output.

 

[RyanW]

Thanks for the info Idiot.

In the new age of digital controls and having a true 4-channel headunit, this should not be an issue. Would there be any other reason that I should not use a larger value resistor here? It seems pretty reasonable to say that, within reason, a higher load impedance will be better.

Can anyone recommend a good signal level audio transformer? I have checked out digikey, but there doesn't seem to be much available. Many state the freq. response as 50Hz to 10khz... not very good. I'm thinking a small transformer with a ~5:1 ratio would do the trick. Are there any North American distributors that might have a better selection?

 

[Perry Babin]

Jensen may have something:
http://www.jensen-transformers.com/

Before you spend too much money on transformers, you should determine if the sound quality of your current head unit is acceptable. At least do a frequency sweep to see if the output has a flat response or has equalization built in (to compensate for OEM speakers with poor response).

 

[RyanW]

Well, one thing is certainly clear... higher quality transformers are very expensive. At $100 - $200 per channel, it is definately not economical. The cheap $10 transformers have very poor frequency response. Looks like the most logical thing to do is to tap out the unamplified signal... as you guys have suggested. I'll have to hit those Mazda forums before re-inventing the wheel.

 

[RyanW]

I've spent a bit of time on the Mazda3 forum and learned a few things.

1. I obtained a datasheet for the IC amp, so I now know which pins carry the pre-amplified signal.

2. I found a person who had tapped the pre-amplified signal from his Mazda 6 headunit. He forwarded me pdf instructions c/w pictures.

For the small effort required, I think that this is the best way to achieve good SQ with a stock headunit. Why screw around with overpriced LOC's when you can, for ~$4.00, bypass the dirty cheap IC amplifier chip and send real signal level to a good amp.

 

[Perry Babin]

Did he measure the output voltage at the point where he tapped into the circuit?

Did he tap off of the inputs of the amplifier IC?

 

[RyanW]

Perry,

Yeah, essentially it looks like he tapped off of the traces leading up to the IC amp chip. He cut a couple traces and replaced the coupling caps with 10uF bipolars. He did not make any mention of the signal level... but I presume that it worked for him without any signal amplifier. I have e-mailed to ask if he knows what the pre-amplified signal level is. If it is low, then I will use a couple of OPA2134's to boost it a bit.

The only part I will need to research is creating the artificial earth for the op-amps. Rod Elliot has posted schematics and a pretty good article on how to do this.

It would be nice to find a schematic of the headunit... I wonder if the signal could go higher than what is required to push the power-amp into clipping. (ie. If the power-amp is disconnected maybe just cranking the volume control higher would push the pre-amp within it's safe operating limit) Not sure if I am conveying my point correctly; but if the pre-amp is on +-6v rails, then although as it is currently set up it never needs to exceed perhaps 500mV, this doesn't mean that it is not capable of delivering a 2V signal. I mean, they most likely used a 4-channel smd op-amp for the pre-amp. Any op-amp I know of can deliver 2V without issue. Am I out to lunch?

Ryan