Musical Fidelity A2 preamp AD797 oscillate

[sviru]

Hi guys. I have made pcb which made possible using four ad797 instead of one QUAD Opamp MC33078-D. What improvement in sound quality - great. But there is one problem. When I turn the volume pot close to miniumum there is a BUZZZZ sound in the speakers. When I touch the volume pot BUZZZING sound gets noticeably smaller. Volume pot is a metal one.

This really bothers me. AD797 gets hot/warm while working - but not that much so You can't touch it.

With 797 this amp sound the best and I need to o something with this.

Decoupling of 797 is made by two tantalum caps from -V and +V pins. Maybe to far away from the supply pins of the opamp?

Please help. Thnx.

[jaycee]

Do not copy the MF preamp. It is a VERY bad design.
The chip is probably oscillating because at low volumes, the gain is too low for the circuit to be stable. Eventually the pot will go bad from DC across it's track.

See this page for more info: http://www.mhennessy1.f9.co.uk/mf_a1/technical.htm

Mark has also made a replacement preamp circuit for the A1 which is far better. He uses an NE5532 but you should be able to adapt it for 2xAD797 if you wish.

AD797 needs high frequency decoupling, try some good 100nF ceramics as close to the AD797 as possible - perhaps fit them beneath the pcb soldered to the pins directly.

[sviru]

Yes this is true. The design od the preamp is no the best one - dc at pot. But after replacing old QUAD opamp the difference is HUGE. I really would like to stay with this design because I don't want to mess inside right now. I'll try to add 100nF as You mentioned. Is there a way to change gain at this circuit when pot is at minimum?

[jaycee]

Not in the configuration used - if you tried to do that, you would find your volume control does not go to zero!

Another alternative might be to bypass it totally and wire the volume control up as an attenuator only. If you only use high output sources such as a cd player or a PC, this would work OK.

[sviru]

Thanks. I generaly use PC and CD. So It could do the job. In my expirience passive preamps always make sound thinner IMHO. That is why I want to keep it that way - by changing opamps I can change the sound in a way I want. Thanks again.

PS. Do you think that changing 47pF caps from ceramic to Silver Mica and resiostors in signal path to something more refined is a good ideaand worth a try?

[sviru]

After adding 0,22uf ath legs of AD797 nothing changed even a little bit. Buziing sound is still there. With AD823 - everything is ok but the soun quality is much worse. Lest transparent, poor resolution etc. Maybe I'll mix AD797 with OPA627 at the output.

[tiefbassuebertr]

the so called "BUZZZZ sound" in the speakers is in most cases oscillation effect, often in the MHz aera. That must be check in any case with an oscilloscope. It is also to do by the pos. and negative supply rails for te OP AMP device. A second reason are residual dc components at the volume control pins. The so called "BUZZZZ sound" is audible in this case only during the turn at the volume control. Then it is more easy to find solution.
In this cases you must first replace the electrolytic capacitors between volume control output and power amplifier input (mostly between 1uF and 4,7uF) through unipolare MKS, MKC or MKH versions (e. g. Siemens or WIMA) and secondly you must choise an OP AMP with jFETs (by AD797 you can introduce additional jFETs).
To save time for troubleshooting I introduce by A1 and AVALON (A200) independend external power supplies for the OP-AMPs in the line- and phono stage. For connecting this devices the brand "binder" is a good solution - see
http://www.binder-connector.de/ximag...8_713810mm.pdf
I upload for you my back engeneered schematic from A1 - I think, the different to your A2 isn't very large by the line amplifier section.

[sviru]

Thanks. The cap before the poweramp is already MKP right now. It seems like the AD797 at the output is not a way to go in here.

I'll try to find something as good as 797. Someone told me to change the resistor 4.7k after the first opamp to 1k to change gain.

Is is correct? If I change this resistor to 1k - does it change gain of the second opamp?

The other idea was to put in parrarel a cap 47p with the pot.

Confused.

[tiefbassuebertr]

If you reduce the resistor behind the first stage to 1K, the max possible gain factor of follow stage is max. 50 instead 10 - I don't recommended this.

an another approach is to remove the complete line amp section (so that you have only passive preamp function). If you have done, you must use GND-ref. for volume control. Therefore you must connected the output of tape monitor switch to the input of volume control. The output of volume control goes to the input caps of power amp. The third connection of volume control goes to GND (for both channels). If the "BUZZZZ sound" is still present, then the reason is by the power amp section or you have a unwanted GND loop.

[sviru]

Passive preamp is a option here but I would really stick here with this one. Can you tell me why increasing gain in this circuit is not recommended? Maybe something like 2-3k would do the job?