Platenspeler.Com |
 |
Platenspeler.Com |
|
Some time ago I received from Sodejuu a pair of AD797 OpAmps especially for this project. And since I'm not a die-hard do-it-yourself fanatic it took far too long to build something together and I passed all my deadlines. Well it will finally happen: I'm going to build my own phono Equaliser based on OpAmps.
Let's first start with looking at the Datasheet of the AD797, in order to get an idea of the do's and don'ts of this chip. Initially I was mainly interested in the power supply requirements and the sensitivity of the input.
Fortunately at first sight the AD797 does not seem to be a very difficult animal: The power supply voltage may vary between the +/- 5 and +/- 15 Volts and that offers sufficient possibilities. The AD797 is according to the datasheet an ideal partner for Microphone, IR and even Sonar applications. Well I have no plans for testing these applications but it is good to know that the chip has very good specifications.
I'm not at all sure that this project will be successful, but read along cause I'm writing down my experience as this project makes progress.
As indicated on the first page, I started with the following design:
The Powersupply rails for V+ (pin7) and V- (pin4) are not drawn in the figure, even as the bias correction (pin8) which is present in some form for some OpAmps. This schema should work for not just the AD797 but several (serious) OpAmps. Whether it works for the AD797 remains to be seen (heard).
It is my experience with DIY projects that once built many projects will be modified several times before the builder is satisfied with the final result. Therefore I will build a prototype first (modular setup if possible) so that I may alter it several times. I realize that in effect I will have finished the amp by then but finding a housing and connecting it to the power supply of the GeenKloon will be the final challenge.
Well, for the time being I put together a separate power supply for the GainPre with the notion that I can always connect the GainPre to the powersupply of the GeenKloon at a later stage. The datasheets of the AD797 mention a Voltage rating of 5-18 Volts but most designs mention a Vcc of 15 Volts and I decided to stick to this value as well.
Meanwhile I'm so satisfied with the result, that I built the power supply as shown in the figure above. And I used a small aluminum case (same series from Conrad as I used for tweety and loekie) and a XLR connection between power supply and amplifier. And below you find a picture of its guts when finished.
For the construction I use epoxy board which makes it easy to handle the AD797. Free wiring is probably not the best way to built the amp. Well, with the board, resistors and capacitors are easy to layout between the chips (and so are power rails etc.). I have decided to give the components minimum space which will allow later upgrades. Wiring will be pure fine silver PTFE tubing isolated.
Scanning through the datasheets I read about the OpAmps sensitivity to static discharge. Also I read about other connectors on the amp chip that may be used to further minimize the distortion and to enhance bandwidth.
The Prototype phono amp is under construction. I first finished one channel and hooked it up. To be honest, I did not have very high expectations of the result. And indeed the first listening impression was not very satisfactory. After studying the schematic I concluded that C1 with an original value of 100nF was either not well connected or not the right value because I hear a lot of high frequencies and very little bass. I was right on both and after replacing the cap with a value of 22nF and changing the connections the nett gain of this one Gainpre channel was equal to the other channel running from Loekie and McStep.
A second listening experience proved that I was on the right track and I decided to build the other channel too (and use the same housing for the amp as I did for the power supply). In the protoamp I used AD797 for both stages of the amplifier.
I've built the other channel of the pre amp over the last few days and started
testing 
today
on one of my amplifiers. Initial tests showed that the gain was not equal for
both channels and it took me quite a while before I figured out what happened:
The local electronics shop had given me one 100 Ohms resistor instead of the
1000 Ohms resistor used in the second stage.
I checked most of the resistor values before soldering them on the board, but
this one must have escaped my attention. Apart from making mistakes with the
values these checks are useful for matching resistor values as well.
The lower channel on the phono is the first channel I finished. Clearly to see is how for the second channel I used a better layout and chip sockets etc. At a later point in time I modified the first channel to sockets as well.
Meanwhile I'm still not satisfied with the behaviour of the AD797 for the input. It clearly is a good sounding OpAmp but it's not stable for MC (low impedance) type of signals. I decided to change the design with respect to the first amplifier stage and subsitute the AD797 for a OPA637 (BurrBrown/TI).
Wow, this was a good decision. The amp is stable now and the sound of the BurrBrown is very good as well. However, noise of the OPA637 is higher and normally an OPA627/637 is used in the last amp section and not attached to some MC cartridge.
IN any case it tells me that the AD797 probably does not like too high an impedance
on it's inputs. This is definately something to remember for the next version
(see phonoclone project)
The sound of the finished pre-amp is quite alright. I'll leave it on 7*24 over the next days to give it a break-in (not connected to power amp).
I do think that the amp is too noisy, so I should experiment with other input impedance settings as well.
| << Page 1: Introduction | Page 3 : Listening>> |