Grundig FineArts A-9000 Amplifier
For an updated page goto German Version
About the FineArts A-9000 amplifier
The symmetrical push-pull circuit of the transistor amplifier stage of the unit is found in almost exactly the same form in equipment from Marantz, Yamaha and others. Nevertheless, my Yamaha AX 630 had a very poor sound with this cuircuit, while the Fine Arts Grundig is convincing. The secret lies is in the sophisticated pre-amplifier stages of the A-9000 which are carefully designed and use discrete parts.
The sound of the fine arts A-9000 is sovereign, transparent and authentic. With
good audio sources and good speakers you get the feeling of musicians
playing life in your room.
The Fine Arts makes a very good performance with the ME25 Geithain speakers that are a spin-off of studio speakers. But the ampilfier has to be modified to deal with that 4 Ohms speakers.
- very clear and intuitive user concept (I have seen nothing comparable)
- Good design, very visually appealing, easy to operate even in the dark, pleasantly soft warm lighting
- 7 inputs (!), independent source selection for a second output
- Very good technical data
|Output power into 8 ohms||2× 120 W Sinus|
|THD at 1 kHz, P-1dB 1 kHz, P -1 dB||<0,005%|
|Intermodulation 1 kHz, P -1 dB||<0,015%|
|TIM 1 kHz, P -1dB||<0,01%|
|Damping Factor 1 kHz into 8 ohms||150|
|Signal to Noise ratio DIN CD (120 W, 50 mW)||>95 dB, 74 dB|
|Power Bandwidth||>5 Hz … 80 kHz|
|Frequency response||7 Hz … 100 kHz|
|Stereo Crosstalk||>70 dB|
Technical data of the Grundig Fine Arts A-9000
Cons: poor quality headphone output (via a resistor divider at the amplifier output)
Adapt FineArts A9000 to run with 230 V
Because of the increase of the power supply voltage in Europe from former 220 V to 230 V (today) the integrated driver STK3102 (Q719) can die. His power supply voltage increases from barely tolerable 58 V to 60.5 V, which is above the authorized the maximum voltage of 60 V. Also it is becomming to hot.
Therefore you have to replace the Hybrid IC STK3102-IV with STK3152III.
This type provides the same sound quality, but it tolerates higher voltages and currents.
For better cooling also add a black fin heat sink using thermal grease which can be mounted at the aluminum heat sink that is already present (and very thin). (Fig)
Additional heat sink on the STK3152.
If the original hybrid circuit is already broken, sometimes the subsequent driver and power transistors may also be burned out due to the damage.
In principle, it is enough to replace them with the same types of transistors.
Warning: Even in Germany you can catch Toshiba transistors fakes. They blow even if no load was applied.
Power transistors of the A-9000
That's why I opted for the safer side and ordered replacement transistors that could handle higher voltage and higher currents. An advantage: they also become less warm.
This results in the following exchange recommendations:
|Transistor||Original type||Replace with|
|Q 711, 712, 715, 716||2SC3182 (NPN, 140 V, 10 A, 100 W, 30 MHz)||2SC5200 (NPN, 230 V, 15 A, 150 W, 30 MHz)|
|Q 713, 714, 717, 718||2SA1265 (PNP, 140 V, 10 A, 100 W, 30 MHz)||2SA1943 (PNP, 230 V, 15 A, 150 W, 30 MHz)|
|Driver stage Q719||STK3102-IV (±60 V, 100 W, 8 Ohm, 0,005 %)||STK3152III (±72 V, 150 W, 8 Ohm, 0,005 %)|
|Additional heat sink||–||yes, mount to the heatsink of Q719|
Original transistors and their replacement models to adapt to 4 ohm load
Run FineArts A9000 with 4 ohm speakers
The amplifier can easily handle 4 ohm speakers. It is designed to do that
The data sheet specifies 8 ohm loudspeakers. But the FineArts A9000 can simultaneously operate two pairs of such 8 ohm speakers (switchable via the front panel as Speakers A and B).
The company FineArtsOnly reported that the amplifier can even handle two pairs of 4 ohms in parallel, as long as you do not run the amplifier with the highest possible volume.
Causes of defects, vulnerabilities
The biggest problem after 20 years of operation were just the switches from ALPS.
When switching the audio source with the source selector knob or with the front buttons there are dropouts or distortions. Wigging the switches helps temporarily. (Solder joints were okay)
Potential failure #1: SPEC121300 from ALPS, unfortunately it can not be fixed
The first major flaw are the mechanical ALPS SPEC switches which tend to corrosion and fatigue (photo above). Disassembly is not possible. When trying to clean the tiny switches from the front panel small components get a loss of adjustment so that they were not usable anymore. These micro switches are definitely not to put back together.
Temporary solution: In the front panel you can exchange a damaged switch
with one of those switches that are not used frequently. At my unit the mono switch and the switch for TAPE1
died first. I have replaced them with the
switch and the
muting switch which were fully functional and which were never used by me.
Also you can solder a jumper on the contacts that are normally closed.
Some big distributors still have the original switches: Search for ALPS SPEC121300 or ALPS SPEC121100.
Repairing ALPS source selector
Update 2013: Some years after the restauration of the switch, which is described in the following, the source selector switch made problems agian. There were drop outs und distortions. So I made a new attempt using
Teslanol Kontakt- und Tunerspray which helped a lot.
The source switch can be dismouldert without removing the pre-amplifier card from the housing. Mind the two additional pads where the housing of the switch is moldert. Than it easily can be removed. You have to disassemble the switch. Then you can clean the contacts by help of a Q-tip using the spray.
The source selector switch from ALPS is prone to corrosion. Fortunately, you can unsolder this switch, remove it, and thus you get access to the corrided contact rails. What I saw then took my breath away.
Potential failure #2: Source selector switch from ALPS with non-enclosed contacts
Purification of the ALPS source selector works like this:
- Get some special oil like the "Wellenschalteröl" from the former GDR. Put a matchstick into the oil. Then dive on the contact surfaces along, gently rub when corrosion is strong.
- Clean switch using a dish detergent and water.
- In a glass or porcelain pot with hot water add 1-2 teaspoons of soda
(do not use baking soda, which does nothing). Wrap loosely
the switch into aluminum foil and insert it. Place the pot into
another pot with boiling water. Now it comes to chemical processes in
which silver sulfide / silver disappears. After an hour the silver
should shine again and large parts of the aluminum foil are dissolved.
Caution: Soda is corrosive and irritates the mucous membranes. Wear eye goggles, and do not inhale fumes.
- Dry the switch, for example using a hair dryer.
- Clean the contact surfaces with a thin shaft of switch oil (Wellenschalteröl). This time it's better to take a cotton swab to prevent that splinters remain.
The same ALPS switch before and after cleaning.
After some years the source selector OF THE Grundig A9000 again had bad contact. I reader suggested using a contact spray (Teslanol® T6-OSZILLIN) which I applied without dismounting the source selector a year ago. It helped a lot, but sometimes you still have to switch on and off when using the outer positions (AUX input).
Other sources of trouble
The ALPS pots for bass, treble and balance produced sometimes slight scratches. Some years ago I carefully opened them and wiped off the the interconnects with a small amount of the red switch oil mentioned above. Since then they work smoothly.
Works wonders - red wave switch oil (rotes Wellenschalteröl)
Resistor R805 and R801 1.2 Ohm / 0.5 W
In both channels these resistors became suddenly high resistance independently of each other. I could not locate any trigger or other damage that caused this. Replace both resistors.
Questions and answers
Is an adaptation / modification of the emitter resistance required? Must the quiescent current be set to a different value?
No, it's enough to replace the transistors, since the relevant data (with exception of the total power dissipation and current) match with the original types. The quiscent current should be the same as stated in the circuit diagram. Adjust as follows:
Left TP1 (-) TP2 (+)
Right TP3 (-) TP4 (+)
Adjustment: Connect digital voltmeter at the measuring points
After switching on the device set the following values:
30 sec after switching on 22...23 mV
1 min after switching on 24...25 mV
In the heated state run the device for about 10 minutes without signal and then match to 28 mV.
The emitter resistors are used as protective resistors. They burn at about 3 A continuous current through.
Does the sound change?
The final stage of the A9000 starts with an operational amplifier that gets an over-all-feedback loop signal from the output back to its input. This essentially determines the sound and it also explains the good audio values that are better than the behavior of a pure AB output stage.
If loudness is set to active, the volume before the output stage has to be halved. 4-Ohm speakers produce a double output power with the same output voltage and thus they are twice as loud (it also depends on the efficiency of the speakers).
When loudness is witched off I feel that nothing is changing in sound.
However, great influence on the sound of the A9000 have corrided contacts of ALPS switches and eventually also the pots. They can produce distortion and dropouts.
WARNING: If the volume pot is at 12-o clock, the amplifier can overdrive when receiving very loud bass signals (subwoofer signals from DVD). That's why I suggest halving the amplitude of the driving audio signal that is feeding the power stage.
How long do you run the A 9000 with the changed components?
The modified A 9000 runs since 2004. Previously it was once blown while running with 4-ohm speakers at high volume STK and power transistors).
Can I use the old thermal pads? Where can I get thermal grease?
The original thermal pads can be reused. Thermal paste is available from the electronics store or computer store.
How do I calculate the power loss?
The following calculation assumes a proper operation of the amplifier, ie, input signals within the proposed range. Provided is further, that the built-in speaker protection circuit works so that no DC voltage can establish.
First, we calculate the voltages generated by the amplifier at the initial maximum power (120 W RMS power into 8 ohms).
It is Û = 1,44 × SQRT (120 W × 8 Ohm) = 44 V peak voltage at the output.
While the current flowing is Î = Û / R = 5,5 A into 8 ohms.
With 4 ohms load you get 11 A peak current.
In the output ampifier the last stage has two transistors that are connected in parallel, which halves the current. The peak current per transistor is thus about 5.5 A at 4 ohms load.
The maximum power dissipation in the transistors is approximately with sinus signals
PT ≈ 0,1 × Ub2 / RL (according to Tietze/Schenk: Halbleiterschaltungstechnik).
PT = 45 W into 8 ohms, divided into two transistors
PT = 90 W into 4 ohms, divided into two transistors