12 replies to this topic

[Rex Dart]

So I've been working on a A/V mod for my model 1 Sega Genesis. I have no formal electronics education, but I've picked up a little bit during 20 years of general fooling around, kit building, and modding.

I did a little test, soldered some RCA jacks to pins 1 and 8 of the CXA 1034 audio chip, and behold... I got some relatively quiet audio. I noticed on the model 1's schematics that the two channels are normally combined on the mobo, and sent to a LM385 op-amp. So I got this NTE987 quad op-amp, supplied it with 5V & Ground connections, wired up the CXA's pins 1 and 8 to a couple of the non-inverting inputs, put audio jacks on the output pins, and... nothing. No audio comes out.

Am I doing something wrong or is the chip defective? Any help is appreciated greatly.

Edited by Rex Dart, Mon Aug 22, 2011 5:35 PM.

[Stephen Moss]

Assuming you have not damaged the CXA 1034 it sounds like you may not have applied feedback to the amplifier circuit, look at the non inverting amplifer section of this page http://en.wikipedia....er_applications

You should place a 100 resistor between output of the amplifer and the RCA connector for the amplifers safety. As a quick test you could try R1 = 1000 ohms and R2 = 5000 ohms (or any value that produce a 5:1 ration) to start with, you can then increase or decrease the ratio from there if you want it louder or quieter. Alternatively use a 10000 ohm variable resistor for R2 then you can adjust the gain (Volume) but turning a knob.

[Rex Dart]

Stephen Moss, on Tue Aug 23, 2011 5:54 AM, said:

...sounds like you may not have applied feedback to the amplifier circuit, look at the non inverting amplifer section of this page http://en.wikipedia....er_applications

Thanks for the info!

I think I get it... so with this being a quad package, there'll be an R1 between the ground and each inverting input, and an R2 (approximately 5x the value of R1) between each output and its corresponding inverting input? I'll have to try that later. I don't think I've damaged the 1034; it still gives its quiet audio when hooked up directly.
Thanks again.

Edited by Rex Dart, Tue Aug 23, 2011 9:33 AM.

[save2600]

I would have just passively tapped into the headphone output and turned up the volume on the TV if need be. Or better yet, simply run a line out of the headphone jack like I'm doing now. You get the added benefit of keeping its sliding pot clean with use.

+1 for the industrious ambition though!

Edited by save2600, Tue Aug 23, 2011 9:48 AM.

[Rex Dart]

save2600, on Tue Aug 23, 2011 9:43 AM, said:

I would have just passively tapped into the headphone output and turned up the volume on the TV if need be. Or better yet, simply run a line out of the headphone jack like I'm doing now. You get the added benefit of keeping its sliding pot clean with use.

Well, I've heard that the parts used in the volume slider area weren't the best, and you get better-sounding output directly from the chip. Is that not the case?

[save2600]

Rex Dart, on Tue Aug 23, 2011 9:56 AM, said:

Well, I've heard that the parts used in the volume slider area weren't the best, and you get better-sounding output directly from the chip. Is that not the case?

No, I'm sure there's a little better performance doing what you are. Just have to weigh whether it's really worth going through all the trouble is all. : )

[Rex Dart]

save2600, on Tue Aug 23, 2011 10:05 AM, said:

No, I'm sure there's a little better performance doing what you are. Just have to weigh whether it's really worth going through all the trouble is all. : )

Man, finding the parts, lining up & drilling the holes for the jacks... that's the difficult part for me. The soldering's the fun part

[Rex Dart]

Here's the schematic I was looking at; you can see how their lone op-amp's used right about in the center of the diagram.

[thomas3120]

I just wired my audio to Pin 1 and pin 8 to the Sony CXA1034P. Sound is good (not quiet) ...not sure if it's the same for different revisions.... This is on a Model 1 VA6
Is yours quiet on both L & R channels or just on 1 channel?

t

[Rex Dart]

It's quiet on both. Definitely about 1/2 to 1/3 the volume of my CDX and Genesis 3.
I'm not entirely sure but I think I have a VA6, too. PCB has both VA5 and VA6 printed on it. Has TMSS, no HD Graphics text, and has the space for the EXT port but no jack soldered.

[Stephen Moss]

Rex Dart, on Tue Aug 23, 2011 9:33 AM, said:

Stephen Moss, on Tue Aug 23, 2011 5:54 AM, said:

...sounds like you may not have applied feedback to the amplifier circuit, look at the non inverting amplifer section of this page http://en.wikipedia....er_applications

Thanks for the info!

I think I get it... so with this being a quad package, there'll be an R1 between the ground and each inverting input, and an R2 (approximately 5x the value of R1) between each output and its corresponding inverting input? I'll have to try that later. I don't think I've damaged the 1034; it still gives its quiet audio when hooked up directly.
Thanks again.

That is correct although you only have to put the gain reistors on the two amplifiers you are using.

Having looked at the curcuit diagram I see you are taking the single from the headphone circuit, therefore the quite sound may be due to the amplifer not having a large enough current output to drive the load you are connecting it to. Thus it is possible you will not need to amplify the signal, just buffer it through an amplifier with a larger current output so for the other two amplifiers in your quad amplifier you might want to connect their outputs to their inverting inputs (unity gain buffer) and their non inverting input to the output of the 1034.
You can then use the output of the buffers as the inputs for the two amplifers with gain, that way you can test both options.

[Rex Dart]

Stephen Moss, on Thu Aug 25, 2011 5:17 AM, said:

Having looked at the curcuit diagram I see you are taking the single from the headphone circuit, therefore the quite sound may be due to the amplifer not having a large enough current output to drive the load you are connecting it to. Thus it is possible you will not need to amplify the signal, just buffer it through an amplifier with a larger current output so for the other two amplifiers in your quad amplifier you might want to connect their outputs to their inverting inputs (unity gain buffer) and their non inverting input to the output of the 1034.
You can then use the output of the buffers as the inputs for the two amplifers with gain, that way you can test both options.

Thanks again, man. Looks like I have more research to do; I don't really understand the difference between amplifying vs buffering through an amplifier, or unity gain buffering... I'll look it up though. Thanks for giving me a starting point.

[Bryan]

A couple things...

When using a single supply, you need to remember to float your input signal between the two supply voltages (0V, 5V). It looks like this is already done for you on pins 1 and 8 of IC12, so tie those two inputs to the non-inverting (+) inputs of two of your op-amps and choose your overall gain according to the following formula:

http://webpages.ursi...uits/node5.html (See figure 25)

Remember, an opamp will not work without both inputs being connected since an opamp amplifies a signal derived from the difference between the two input pins. Just connecting the output to the inverting input will give the same voltage out as you put in on the non-inverting input (the opamp will produce an output that keeps both input pins at the same potential- called a 'follower'). A gain of 2-5 is probably about what you want but you can go as high as you want until you exceed the limitations of a 5V supply. Try to keep the values between 1K ohms and 10K ohms for the two resistors, and use IC12 pin 5 for the R1 "ground" connection from the inverting input. Just remember that opamps typically run on 'dual rails' (like +12V, -12V) so ground sits between them at 0V. When we don't have dual rails but instead are using 0V and 5V, we need a "fake" ground around 2.5V to connect to. IC12 pin 5 appears to be a voltage divider output for just such a purpose.

Also, unused opamps can do all kinds of crazy things if not connected in some way. I would connect IC12's pin 5 to both unused + inputs and connect the unused outputs to their - input.