SX4 OEM HU: By pass surgery for Amp section

omishra · 5th August 2008, 10:31

Quote:

Originally Posted by greenhorn

PS: another suggestion. Have your thought aboug upgrading the stuff currently in the signal path ( op amps, capacitors etc ) ?

No I am not thinking to upgrade that. Reason for this is that all semiconductors are SMD and quality of PCB is pathetic to work.

Again I could not get cap in market (at least here in Pune) which are better than OEM parts.

Making further changes in this direction, it will achieve better sonic results but considering the listening environment I think it will make less effect in "InCAR" environment.

What you say?

greenhorn · 5th August 2008, 10:32

hmm quite true. Like we've all said , a speaker upgrade + some damping would be the best upgrade at this point

DerAlte · 5th August 2008, 10:43

Super, Mishra-ji, this is a stupendous effort! Your DIY effort will inspire a couple of generations of DIY enthusiasts.

@green, changing the op-amps won't make much of a diff, since the operation is in a narrow band of performance of the op-amp. Changing the series coupling capacitance may have a detrimental effect - larger the cap, more the self-generated noise and EMI-induced signal pass-through. 0.22 used to be the most optimum for audio spectrum.

omishra · 5th August 2008, 10:50

Quote:

Originally Posted by greenhorn

hmm quite true. Like we've all said , a speaker upgrade + some damping would be the best upgrade at this point

Quote:

Originally Posted by DerAlte

0.22 used to be the most optimum for audio spectrum.

Taken a note of it Sir. Yes I need to clean the signal path because new AMP is sensitive enough to pickup constant hiss generated at signal path. I need to supress that. Though that is heard during silence between two tracks or low volumes.

greenhorn · 5th August 2008, 10:58

@ DerAlte I thought so because he had issues with the residual noise. I've often seen op amps like the LM 324 used in such applications. Never been able to find a better pin compatible replacements locally though.

Though i doubt if any op amps are in omishra's current signal chain since he's taking the output straight off the DAC.

omishra · 5th August 2008, 11:09

Quote:

Originally Posted by greenhorn

@ DerAlte I thought so because he had issues with the residual noise. I've often seen op amps like the LM 324 used in such applications. Never been able to find a better pin compatible replacements locally though.

Though i doubt if any op amps are in omishra's current signal chain since he's taking the output straight off the DAC.

There is no OP AMP in signal path though one electrolyte Cap is there, don't remember the value. Pre-Out has transistorized signal-amp which deteroites SQ, that is confirmed by me.

Do I need to add some Lm324 pre-amp kind of thing to get output of DAC over 1 feet signal wire?
I want to supress residual noise, audio level is now adequate.

greenhorn · 5th August 2008, 11:13

Anything other than the LM 324

- its shouldnt be allowed to stray into an audio path, ever.

Is it just noise or interference ? You said that the noise disappears when you use another source. So its from the HU i guess. Dont think increasing the pre out levels , especially for such short distances will help a lot.

After thinking a bit, try anyway. maybe the noise is there because the DAC output is high impedance

anujmishra · 5th August 2008, 11:18

One word I will say, Great! How much knowledge Team-BHP members carry ?

This is one of the fine and complex DIY performed. I know satisfaction after doing DIY is immeasurable.

mav2000 · 5th August 2008, 11:41

Super stuff and one of the best DIY's I have seen and something that probably thousands of SX4 users would love to do. Keep going and best of luck for the next stage.

mohang_j · 5th August 2008, 12:16

Quote:

Originally Posted by greenhorn

Anything other than the LM 324

I have experienced TL084 to have better s/n ratio.-May be i am wrong

Quote:

Originally Posted by ommishra

It could be 22W @ 14.4W and 1% THD for 4 ohm speaker. But I can get double of that if I can hook 2 ohm load.

Is there any reason you have not selected TDA73xx AMP.What about trying Tantalum capacitors in signal path.

Bass&Trouble · 5th August 2008, 12:33

OMG, thats some serious DIY! Great going, Mishraji!

A lot of it is full "bouncer" to me too. But its great to see that you have unearthed and rectified the exact drawbacks that we experience with nearly all OEM HUs. And to think that a lot of users are reluctant to change their OE Hu and some also claim that the sound quality is "quite good"! This will serve as an excellent reference thread. Now, if only I could have heard this.

Do post any further developments. All the best! cheers:

omishra · 5th August 2008, 12:45

Quote:

Originally Posted by greenhorn

After thinking a bit, try anyway. maybe the noise is there because the DAC output is high impedance

BANG.. You are exactly there. This needs matching.

Ohh man how can I forget this?

My Amp is having input impedance 50K ohm, original chip was having 90K. between DAC and OEM chip there was 100K resistor. Now I have connected it directly after DAC's DC blocking cap. I need to match impedances.

Thanks for pointing it out.

DerAlte · 5th August 2008, 13:37

Quote:

Originally Posted by omishra

... Pre-Out has transistorized signal-amp which deteroites SQ, that is confirmed by me. ...
I want to supress residual noise, audio level is now adequate.

Why should the transistorized intermediate amp deteriorate SQ? I think you are being empirical - there are no data points here.

Quote:

Originally Posted by omishra

... My Amp is having input impedance 50K ohm, original chip was having 90K. between DAC and OEM chip there was 100K resistor. ... I need to match impedances.

This reminds me of the Japanese craze of straightening hair! How / why in heavens are you trying to match impedances without ANY means of signal or impedance measurement? Not a good idea to remove components from the signal path without examining their role.
Noise is affecting your signal since early in the signal path the SNR is bad - not because noise is more, but because signal is less. What was OK with the Toshiba chip (there is an op-amp in the input stage, whose gain you have ignored) is not OK without it.

greenhorn · 5th August 2008, 13:43

omisra, do you remember the name of the DAC chip ?

rather than adding a resistor again in series. Adding an op amp stage like you mentioned to buffer (and possibly raise) the DAC output levels would be the better way to go. Remember to use a good op amp.

I dont think impedance matching will be an issue since the amp board you built must have a high impedance ( cant say for sure. couldnt find the circuit diagram on the link you gave - where does the input signal go first? I saw a 16 pin IC . which one is that/any idea what it does) . The impedance matching is not an issue , but high impedance outputs tend to be prone to interference - that's why they are buffered by an op amp stage before they are sent off board

esteem_lover · 5th August 2008, 13:48

This is an awesome mod. Congrats mishra. This option could come in very useful for many. Thanks for sharing.

5th August 2008, 10:32	#17
greenhorn Senior - BHPian Join Date: Apr 2007 Location: KL-01 Posts: 7,745 Thanked: 4,401 Times View My Garage	hmm quite true. Like we've all said , a speaker upgrade + some damping would be the best upgrade at this point
	() Thanks

5th August 2008, 10:43	#18
DerAlte Senior - BHPian Join Date: Mar 2007 Location: Bangalore Posts: 8,069 Thanked: 2,919 Times	Super, Mishra-ji, this is a stupendous effort! Your DIY effort will inspire a couple of generations of DIY enthusiasts. @green, changing the op-amps won't make much of a diff, since the operation is in a narrow band of performance of the op-amp. Changing the series coupling capacitance may have a detrimental effect - larger the cap, more the self-generated noise and EMI-induced signal pass-through. 0.22 used to be the most optimum for audio spectrum.
	() Thanks

5th August 2008, 10:58	#20
greenhorn Senior - BHPian Join Date: Apr 2007 Location: KL-01 Posts: 7,745 Thanked: 4,401 Times View My Garage	@ DerAlte I thought so because he had issues with the residual noise. I've often seen op amps like the LM 324 used in such applications. Never been able to find a better pin compatible replacements locally though. Though i doubt if any op amps are in omishra's current signal chain since he's taking the output straight off the DAC. Last edited by greenhorn : 5th August 2008 at 11:00.
	() Thanks

5th August 2008, 11:13	#22
greenhorn Senior - BHPian Join Date: Apr 2007 Location: KL-01 Posts: 7,745 Thanked: 4,401 Times View My Garage	Anything other than the LM 324 - its shouldnt be allowed to stray into an audio path, ever. Is it just noise or interference ? You said that the noise disappears when you use another source. So its from the HU i guess. Dont think increasing the pre out levels , especially for such short distances will help a lot. After thinking a bit, try anyway. maybe the noise is there because the DAC output is high impedance Last edited by greenhorn : 5th August 2008 at 11:15.
	() Thanks

5th August 2008, 11:18	#23
anujmishra Senior - BHPian Join Date: Jun 2008 Location: Bangalore Posts: 1,290 Thanked: 492 Times View My Garage	One word I will say, Great! How much knowledge Team-BHP members carry ? This is one of the fine and complex DIY performed. I know satisfaction after doing DIY is immeasurable.
	() Thanks

5th August 2008, 11:41	#24
mav2000 BHPian Join Date: Jan 2008 Location: Bangalore Posts: 621 Thanked: 315 Times View My Garage	Super stuff and one of the best DIY's I have seen and something that probably thousands of SX4 users would love to do. Keep going and best of luck for the next stage.
	() Thanks

5th August 2008, 12:33	#26
Bass&Trouble Senior - BHPian Join Date: Mar 2006 Location: Bombay Posts: 2,754 Thanked: 124 Times	OMG, thats some serious DIY! Great going, Mishraji! A lot of it is full "bouncer" to me too. But its great to see that you have unearthed and rectified the exact drawbacks that we experience with nearly all OEM HUs. And to think that a lot of users are reluctant to change their OE Hu and some also claim that the sound quality is "quite good"! This will serve as an excellent reference thread. Now, if only I could have heard this. Do post any further developments. All the best! cheers:
	() Thanks

5th August 2008, 13:43	#29
greenhorn Senior - BHPian Join Date: Apr 2007 Location: KL-01 Posts: 7,745 Thanked: 4,401 Times View My Garage	omisra, do you remember the name of the DAC chip ? rather than adding a resistor again in series. Adding an op amp stage like you mentioned to buffer (and possibly raise) the DAC output levels would be the better way to go. Remember to use a good op amp. I dont think impedance matching will be an issue since the amp board you built must have a high impedance ( cant say for sure. couldnt find the circuit diagram on the link you gave - where does the input signal go first? I saw a 16 pin IC . which one is that/any idea what it does) . The impedance matching is not an issue , but high impedance outputs tend to be prone to interference - that's why they are buffered by an op amp stage before they are sent off board Last edited by greenhorn : 5th August 2008 at 13:52.
	() Thanks

5th August 2008, 13:48	#30
esteem_lover Senior - BHPian Join Date: Jul 2006 Location: Madras/Py Posts: 7,556 Thanked: 502 Times View My Garage	This is an awesome mod. Congrats mishra. This option could come in very useful for many. Thanks for sharing.
	() Thanks