TECHNICS ST GT650 TUNER Mod

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2006
Paid Member
Just for fun, I will now start to mod an inexpensive Technics tuner hoping to bring it´s sound to a higher degree.

My plan includes powering the output analog opamp with a Salas BiB.

As I would like to make it as user friendly as possible (It is a family tuner), I need all the help I can get to retain all it.s functionalities.

First step would be to power IC 302 with the BiB, but I would like to have the BiB controlled by the main controller ship (power on/off/standby).

The 13.7V are produced by the Q701 and Q705 arrangement... can I somehow feed the 13.7volts from the BiB here without compromising functionality ?
 

Attachments

  • sch1.png
    sch1.png
    102.3 KB · Views: 349
  • sch2.png
    sch2.png
    134.1 KB · Views: 338
  • sch3.png
    sch3.png
    105.1 KB · Views: 348
  • sch4.png
    sch4.png
    108.1 KB · Views: 317
  • sch5.png
    sch5.png
    100.6 KB · Views: 297
  • sch6.png
    sch6.png
    115.1 KB · Views: 241
I haven't a clue what a BiB is, so I can't answer the question, but as the sound quality of an FM tuner is largely determined by its IF filter response and its FM demodulator, have you considered tweaking those?

For example, using a length of terminated coaxial cable for frequency to phase conversion in a quadrature detector will work much more linearly than the typical LC-filter-based or ceramic-resonator-based circuits. Still, it is almost never used (except in Studer tuners like the A726) because it is more expensive and heavier.

For the IF filter, you could replace the ceramic filters with low group delay ripple types, unless those are already used of course.
 
Member
Joined 2006
Paid Member
I haven't a clue what a BiB is, so I can't answer the question, but as the sound quality of an FM tuner is largely determined by its IF filter response and its FM demodulator, have you considered tweaking those?

For example, using a length of terminated coaxial cable for frequency to phase conversion in a quadrature detector will work much more linearly than the typical LC-filter-based or ceramic-resonator-based circuits. Still, it is almost never used (except in Studer tuners like the A726) because it is more expensive and heavier.

For the IF filter, you could replace the ceramic filters with low group delay ripple types, unless those are already used of course.

Hi Marcel

Thank you for the input. The BiB is a special shunt psu that usually produces great results when powering analog circuits.

As for the coax mod you are refering to, I would really like to implement it but my knowledge about tuners is scarce.... can you point that out in the circuits from the schematics I posted ? (If you need a pdf I can send it to you by mail)

About the IF filter replacements... are you talking about ceramic caps ?

What is a low group delay ripple type ?

Regards

Ricardo
 
Hi Ricardo,

No, I mean ceramic bandpass filters. I saw in the schematic that your tuner uses ceramic filters, but I couldn't read the type number.

There are many types of those filters. Some have a relatively narrow passband, steep roll-off and a poor group delay characteristic. Others have a wide passband, less steep roll-off and good group delay characteristics.

Narrow filters with a steep roll-off and poor group delay characteristic give more distortion, especially on stereo signals, but also more selectivity. So they give the best overall reception when you want to receive a weak signal while there is a strong undesired station with a frequency close to the weak wanted signal.

When you want to receive a strong desired signal and there is no interferer right next to it, then you can improve the sound quality by going for a wide passband and good group delay.

Could you send me the pdf if it has better resolution than the pictures you posted?

Best regards,
Marcel
 
The Technics ST-G70/90 uses the same MPX and output IC but simpler filtering, still uses that Class AA config.
6.2V offset into the 6554 quad opamp and 5.7 to polarise the electrolytic blocking caps?
 

Attachments

  • technics_st-g70_g90_sm.png
    technics_st-g70_g90_sm.png
    169.3 KB · Views: 152
Regarding the IF filters:
you could replace CF101 and CF102 with filters with a wide bandwidth and a good group delay, for example Murata type SFELF10M7FA0G-B0, see

http://www.murata.com/products/catalog/pdf/p50e.pdf#SFELF

I don't know if this will improve anything though: I haven't found any information about the Panasonic filters RLFFETNG*01L and 02L that are presently used, so I don't know if the Murata parts are better or worse. As the stereo distortion specifications of your tuner are rather good, it could be that it already has wide filters with good group delay characteristics.

On top of that, when you replace the filters, you have to hope that the Panasonic filters are designed for 330 ohm terminating resistance, like the Murata ones. That is rather likely, as 330 ohm is by far the most common terminating resistance for ceramic IF filters.

There is no reason to replace CF103 and CF104, they are only switched in the circuit in the narrow IF mode, which you would normally only switch on when you have bad reception and want a narrow filter. As the filters come in different centre frequency classes, make sure the new CF101 and CF102 and the old CF103 and CF104 have the same centre frequency.

About the demodulator: IC102, AN7274NS, is your demodulator IC and C112, R110, T101 and T102 are the frequency to phase converting circuit. You would have to replace this circuit with a buffer amplifier, for example an emitter follower, driven by pin 16 of the AN7274NS, a DC blocking capacitor, a thin coaxial cable and a terminating resistor to pin 26, the 5.6 V supply. The end of the cable would be connected to the resistor and to pin 16 of the AN7274NS. I can draw up a proposal for a schematic if you seriously want to try this.

You would have to measure the audio output level of the tuner before doing any modifications. Then make the changes and make the coaxial cable one quarter of a wavelength of 10.7 MHz long. For a normal polyethylene filled cable that boils down to
(1/4)*0.67*(299792458 m/s)/(10.7E6 MHz)~=4.69 m.

Then measure the output signal level and if it is too low, which is likely, replace the coaxial cable with a longer one, but make sure it is still an odd number of quarter wavelengths. For example, if the audio level would be 10 times too small, you could replace the cable with one that is nine times longer.

Actually ideally the length should be 3/4+n wavelenghts, as 1/4+n wavelengths would change the polarity of the audio signal. Of course that can be corrected by adding an inverting stage somewhere in the audio path, and I'm not sure if the orginal tuner has the correct polarity to begin with.

Anyway, you probably would have to make the cable a bit longer than calculated, measure the DC offset between testpoints TP101 and TP102, make the cable a little bit shorter, measure the DC offset again and as such trim the cable length until the DC offset is within spec. You would have to do this adjustment with the tuner properly tuned to an RF input signal (either from a signal generator or a strong radio station with no adjacent interferers).
 
Last edited:
Member
Joined 2006
Paid Member
Thank you for your input.

I will look into the Murata filters. As for the phase converting circuit, it looks like a worthile mod but unfortunately I do not have any measurement equipment now so it would be very difficult to implement.

What I believe I can do is replacing Q701 regulator for something much better. (A shunt reg :))

But how can I implement it retaining the tuner on/off capabilities ? And where can I feed the 13.7V ? (There is a connection between Q701 emitter and Q702 (5.6v reg) that I do not fully understand)

Maybe I should leave these two regs intact and only feed a clean +13.7v DC into the output quad opamp...maybe replace output caps and see...
 
By the way, if you should change the IF filters, readjust the channel separation if you have access to the required equipment. The required equipment is a good FM stereo signal generator (or a very low-power FM transmitter) and something to measure the output level of the channel that should be silent.

You can even measure the output level with an amplifier, loudspeaker and your ears, as you just need to find out at what setting the silent channel is at its most silent. The channel separation can be trimmed with VR301, as indicated on page 14 of the pdf you sent.
 
Member
Joined 2006
Paid Member
That can only be done with the help of a FM stereo signal generator..... but thank you for the tip.... I will keep it for future reference.

Returning to the PSU... I think that the 20V on Q701 collector are obtained from a voltage multiplier (C705 706)... Am I correct ?
 
Hi Ricardo,

I'll get back to the PSU questions later, after looking very carefully at the schematics.

About the measuring equipment needed for my modifications: you definitely need some sort of DC voltmeter for the offset trimming and a takemeasure for measuring the length of coaxial cable. For everything else you can find solutions if you get creative/pragmatic:

-The audio signal level measurements are relative measurements, so you can do them with an audio recording device having a VU or peak meter or a signal waveform display. A computer with a sound card and Audacity will work fine (as long as you make sure no-one touches the recording level setting between the measurements).

-As a stereo signal generator, you could use one of those very low power FM transmitters that people use nowadays to listen to their MP3 player over their car radio, if these are legal in your country and if you have one or can borrow one. Play your favourite Bach cantate over the left channel of the transmitter and adjust the receiver until you hear the least music in the right channel.

-If you don't have one of those, tune in to your favourite FM station and adjust VR301 until you like the stereo imaging best. This probably won't be the theoretically correct point, but who cares as long as you like the sound?

Best regards,
Marcel
 
Thank you for your input.

I will look into the Murata filters. As for the phase converting circuit, it looks like a worthile mod but unfortunately I do not have any measurement equipment now so it would be very difficult to implement.

What I believe I can do is replacing Q701 regulator for something much better. (A shunt reg :))

But how can I implement it retaining the tuner on/off capabilities ? And where can I feed the 13.7V ? (There is a connection between Q701 emitter and Q702 (5.6v reg) that I do not fully understand)

Maybe I should leave these two regs intact and only feed a clean +13.7v DC into the output quad opamp...maybe replace output caps and see...

It looks to me like the collector of Q706 drives switching transistor Q705 which turns on or off the 13.7 V supply. Regulator transistor Q701 then supplies the reference (zener diode) of the 5.6 V regulator through resistor R703. Hence, switching off the 13.7 V automatically switches off the 5.6 V. On top of that, the line regulation of the 5.6 V supply is much better than it would have been had R703 been connected to an unregulated supply, because there is less ripple current flowing through the zener.

So you can disconnect the emitter of Q701 and connect your alternative regulator to everything that is normally connected to the emitter of Q701. To retain the normal on/standby switching function, your regulator has to have an enable input. You can drive an active-low enable input from the collector of Q706 and add a pull-up resistor. For an active-high enable, connect the emitter of Q705 to an always-on supply carrying the proper high level, connect its collector to the enable input and add a pull-down resistor.
 
Hi, I am Carlos from Denmark and I have a Technics ST-G70 tuner. Could you recommned some modification? I've heard to change some capacitor would develope the sound quality a bit. Here is a link, in Germany they made some modification:
Klangtuning für Technics ST-G 90 - Hi-End Tuner
Could you check the picture? I wrote them a letter, but only they asked the following about the capacitors:
"
Take Nichicon Fine Gold.
For the power supply / Elna smoothing power supply it.
But they didn't tell me about the value of the capacitors. (Voltage, F)
Can you help me?

With kind regards,
Carlos
 
Member
Joined 2006
Paid Member
It looks to me like the collector of Q706 drives switching transistor Q705 which turns on or off the 13.7 V supply. Regulator transistor Q701 then supplies the reference (zener diode) of the 5.6 V regulator through resistor R703. Hence, switching off the 13.7 V automatically switches off the 5.6 V. On top of that, the line regulation of the 5.6 V supply is much better than it would have been had R703 been connected to an unregulated supply, because there is less ripple current flowing through the zener.

So you can disconnect the emitter of Q701 and connect your alternative regulator to everything that is normally connected to the emitter of Q701. To retain the normal on/standby switching function, your regulator has to have an enable input. You can drive an active-low enable input from the collector of Q706 and add a pull-up resistor. For an active-high enable, connect the emitter of Q705 to an always-on supply carrying the proper high level, connect its collector to the enable input and add a pull-down resistor.

Hi Marcel
Just finished building the 14v shunt (Salas BiB p2p) and my plan is to power it using an external TX.
To retain on/off capabilities I am thinking using a 12v power relay to switch on the shunt.

I will disconnect the emitter of Q701 and feed there the 14Vdc.

Can I use Q701 emitter output to power the12v relay (that should switch on the secondary of the external tx, just before the rectifier diodes)

I plan to use a PCB Power Relay – G5NB-E (12V coil, 720r).

Regards

Ricardo
 
Member
Joined 2006
Paid Member
Hi Carlos

I do not know if the ST-G 90 is similar to the tuner I am upgrading, so solutions might be different.

I believe that a good psu is very important to any audio equipment so that is where I am starting.

In the link you provided, they alk about replacing psu caps, output caps and output opamp... this gave me "food for the mind" and I am already searching for a good replacement for the AN6554 quad opamp... but that will be second phase only.

For now I will be replacing the basic 14vdc psu with an external shunt, hoping to retain on/off capabilities (this is a family tuner and I need everyone to be able t use it).

Let us know if you have a schematic for your tuner so we can study it.

Regards

Ricardo
 
Hi Marcel
Just finished building the 14v shunt (Salas BiB p2p) and my plan is to power it using an external TX.
To retain on/off capabilities I am thinking using a 12v power relay to switch on the shunt.

I will disconnect the emitter of Q701 and feed there the 14Vdc.

Can I use Q701 emitter output to power the12v relay (that should switch on the secondary of the external tx, just before the rectifier diodes)

I plan to use a PCB Power Relay – G5NB-E (12V coil, 720r).

Regards

Ricardo

Yes, that should do the trick (at least I can't think of any reason why it wouldn't). Maybe it is a good idea to disconnect the electrolytic capacitor at the base of Q701 (C707?) to make the relay switch faster and reduce contact wear. You won't need a diode for clamping the voltage at switch-off, as this is already taken care of by the zener (in forward conduction mode) and Q701.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.