12
« on: December 24, 2009, 11:47:11 AM »
I have done a full tube repro on my A810 about two years ago. I did that same full tube repro mod on my Revox C270 six months before that.
That mod consists of a 23dB single tube flat gain stage built in the tape machine itself, connected to the heads by short cables. The output of that stage is capable of driving a long cable so that the rest of the gain stages and the different EQs can sit in a separate external enclosure. The reason for that approach was mainly the fact that Studer heads don't do well with any significant capacitive loading, so the cables between the heads and the first gain stage need to be short. The output of the first gain stage has a fixed resistive output impedance and can easily drive a long cable without degrading performance.
The mod is permanent, but in order to keep the machine at least fully functional stand alone, the output of the tube gain stage also goes to a 23dB resistive attenuator and is then fed back to the original head preamp. That way I can take full benefit of the tube repro while the VU meters are still calibrated and the machine can still be used stand alone, without the tube repro.
The tube repro worked extremely well on the C270, so I thought: Let's do exactly the same thing on my A810. Only some small frequency curve corrections would have to be made in the gain stage to compensate for the difference in characteristics of the playback head in the A810 compared to the one in the C270. that way I could use the same back end amplification and EQs for both the C270 and the A810 without the need for adjustment.
The first gain stage in the C270 was placed behind the tape controls and counter. There is an unused slot that is reserved for a timecode card if I remember well. Unfortunately there was not such a nice location like that in the A810. Since it is extremely busy inside the A810 there is only one spot with sufficient useful volume, which is on the left side of the middle towards the back. The only way to actually put something there is mounting it to the back plate of the machine. Inconvenient but feasible. That way the head wires could be kept under a feet. Making them longer will compromise the transparency of the sound.
So I built it up that way with the first gain stage built in an aluminum enclosure mounted to the back. After the success with the C270, my expectations of the A810 where very high. So the slap in the face was ever so more painful when I played the first tape. Somewhere between the hum and the noise there was also some music. I must have pulled the preamp out at least four times to double check the performance on the bench, which showed the same performance as the C270 preamp time after time.
To make a long story short, I found out the hard way why the head preamp of the A810 is sitting on the head block in a way that looks like an after thought and why there is so much shielding. The noise levels below the diecast chassis of the A810 are enormous. I could pick up volts over the ground lead of a shorted scope probe. The origin of some of that noise is the switch mode power supply for the 5V bus, but mostly the PWM regulated motor controllers. The designers made a well performing tape tension control, but clearly had no clue about EMI management.
I had to spend days to reduce the switching noise with at least 30dB to get the noise and hum of the preamp to a level that was sufficiently below the noise floor of the tape.
There is a happy end to the story in that the A810 now sounds better than anything that I have had the opportunity to listen to and that it has been my primary playback machine now for the last two years.
The moral of the story is: be very careful with running the un-amplified head signals through the machine. Long cables don?t work all to well with Studer heads, but if you want to do it anyway, my recommendation would be the run the head cables over the diecast chassis in the direction of the output plugs rather than under the diecast chassis.
Show Posts