Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[eddieaudio]

I just listened to the Lawrence Welk LP  "Calcutta" and the liner notes state that the Fairchild 641 mastering system was used.  The pressing is mint and the record sounds great.  Good recording, big wide stereo and a lot of depth - you can hear the room.  Seems like there are two bass guitars, electric on the left and acoustic on the right.  Since the Fairchild was sum and difference, perhaps panned basses might have been less of an issue...

[Thomas W. Bethel]

Paul

... I believe the tube ones were push-pull with 811? output tubes. The big ones with the ceramic heater wire on top. The most common SS are the blue ones.

This is an 811 Tube http://sarris.info/main/files/SV811-3Aspec.pdf

This is an 807 with a plate cap http://sarris.info/main/files/GL807.pdf

Not sure which one you were referring to???

Why today couldn't you use a DIGITAL amplifier for a head amp??? Seem like it would be the best of all. Really low output impedance and lots of power in a very small package with frequency response of DC to light.

[TotalSonic]

eddieaudio wrote on Thu, 31 December 2009 16:43

I just listened to the Lawrence Welk LP  "Calcutta" and the liner notes state that the Fairchild 641 mastering system was used.  The pressing is mint and the record sounds great.  Good recording, big wide stereo and a lot of depth - you can hear the room.  Seems like there are two bass guitars, electric on the left and acoustic on the right.  Since the Fairchild was sum and difference, perhaps panned basses might have been less of an issue...

Panned basses would have presented the same issue of high & wide amounts of vertical movement regardless of whether the cutter head is lateral/vertical at 90degrees like the Fairchild 642 or using the more common left/right at 45degrees like the Westrex 3D/Haeco SC-2 or Neumann SX68/74/82.  

Best regards,
Steve Berson

[Gold]

Thomas W. Bethel wrote on Thu, 31 December 2009 18:19

Really low output impedance and lots of power in a very small package with frequency response of DC to light.

Well you don't want DC and you REALLY don't want light. The possibility of switching noise anywhere near the cutterhead seems like an unnecessary risk.


It's the 807 tube.

[bruno putzeys]

Quote:

couldn't you use a DIGITAL amplifier for a head amp??? Seem like it would be the best of all. Really low output impedance and lots of power in a very small package with frequency response of DC to light.

I get this question mailed directly to me by cutting engineers with surprising regularity. I'll leave aside the usual diatribe about the verifiable nonexistence of something like a digital amplifier (it's at http://www.hypex.nl/docs/allamps%20hypex%20layout.pdf if anyone's interested).

The important question is: what are the advantages to using class D?
*It's efficient
*It's compact

Those are the *only* defining characteristics which cannot be emulated or obtained more easily using non-switching power stages. In all other aspects, class D is a decided burden. Characteristics which are possible, but very difficult to achieve in class D are:
*Low distortion, low output impedance, uncoloured sound etc. Some makes of class D amps have a reputation for good sonic and measured quality. This is not because of class D but in spite of it. It's really hard.
*Absense of a switching residual. A small amount of residual is of no consequence in a loudspeaker, but would you like to cut a 400kHz signal, even a very tiny one, into a groove? I think not.

Characteristics of linear power amps that are impossible to get with switching power stages are:
*Simplicity of design
*Very wide bandwidth
And in the context of cutting
*Very high output impedance
*Ease with which a supplementary feedback loop is added

To make a long story short. Efficient is nice when a million people have their stereos on. Compact is nice when the product goes into your living room. Neither are particularly important when you're cutting records. Each record is cut exactly once, so the heat output is lost only once. And 4u of rack space next to the lathe isn't going to spoil an already geeky interior.

What you do need is:
*A feedback loop from the sense coils, preferably one that doesn't take a seasoned control theoretician two years to work out. A wideband amplifier makes this almost simple.
*Current out (one could work around that).

A class D amp to drive a cutting head solves nothing and only adds more problems.

[Thomas W. Bethel]

Bruno,

Thanks for the very clear and very readable answer to my question. It is professionals like you that make this a GREAT forum! Thanks also to Brad for providing such a GREAT forum!!!

[Gold]

Bruno Putzeys wrote on Thu, 07 January 2010 08:04

*Absense of a switching residual. A small amount of residual is of no consequence in a loudspeaker, but would you like to cut a 400kHz signal, even a very tiny one, into a groove? I think not.

"That smoke doesn't smell like lacquer"

Quote:

Characteristics of linear power amps that are impossible to get with switching power stages are: *Simplicity of design *Very wide bandwidth And in the context of cutting *Very high output impedance *Ease with which a supplementary feedback loop is added

I'm not sure if high output impedance is a typo but it is not necessary for a cutting amp. An SX74 has an impedance of 6 ohm @1K. A plain old audio power amp will work fine.

In the Neumann racks the power (drive) amps are separate from the feedback amps and summed in yet another module.

I just looked on the VG66 feedback amp and the input impedance is 1.8k. It looks like most mic pre's could do the job without modification of the input.

[bruno putzeys]

The switching residual represents a few tens of microwatts into a normal voice coil based transducer, so there's never the risk of heating anything. That the HF should somehow cause thermal problems is one of the toughest misconceptions about class D. This is why it won't matter in a speaker. I just don't think cutting HF into a groove is a good move.

Re output impedance, it's quite normal in motor control systems to drive the motor coil with a current source and take all feedback from a motion sensor. It makes feedback loop design a lot easier. If I had to design a cutter head amp that'd be the first thing I'd try.

[Gold]

Bruno Putzeys wrote on Fri, 08 January 2010 02:52

The switching residual represents a few tens of microwatts into a normal voice coil based transducer, so there's never the risk of heating anything. That the HF should somehow cause thermal problems is one of the toughest misconceptions about class D. This is why it won't matter in a speaker. I just don't think cutting HF into a groove is a good move.

The RIAA curve doesn't have a poll to roll off after 15-20k Hz. So that few microwatts will become smoke.

Quote:

Re output impedance, it's quite normal in motor control systems to drive the motor coil with a current source and take all feedback from a motion sensor. It makes feedback loop design a lot easier. If I had to design a cutter head amp that'd be the first thing I'd try.

I can't analyze the circuits well enough to say whether this is happening or not.

[bruno putzeys]

The RIAA curve is applied before the power amp. Not after. So, the amplitude of the switching residual is not affected by the fact that the signal going into the amp has an RIAA correction applied.

[Gold]

You sir, are correct.

[Geoff Emerick de Fake]

Bruno Putzeys wrote on Fri, 08 January 2010 01:52

Re output impedance, it's quite normal in motor control systems to drive the motor coil with a current source and take all feedback from a motion sensor. It makes feedback loop design a lot easier. If I had to design a cutter head amp that'd be the first thing I'd try.

IIRC, the Neumann SAL74 amp used a brige detector servo configuration. The idea was that an SX74 could be replaced with an older head (such as SX68). I had to do that once because the SX74 had fried. A complete recalibration was indeed needed, but the machine was up and running in a couple of hours. Having the SX74 re-kitted was a matter of months at the time! and it was impossible to have a temporary loaner.

[Gold]

I think you are talking about the circuit breaker not the power amp arrangement. The DC resistance of the cutter head is used as one side of the bridge. If the drive coils heat up the resistance goes up and the breaker trips.

[Geoff Emerick de Fake]

Yes, but I think that, in normal operation, the differential bridge signal is used to servo control of the stylus.

[Gold]

I thought the the circuit breaker was essentially "out of the signal path". The only thing in line with the cutter head is the relay. The rest of the circuit breaker is the detection side chain.  

I looked on the VG66 SI66 schematic and I see the current source fed to the cutter head for detection in the circuit breaker. I don't see anything that goes to the cutter head besides that. I checked the system diagram and saw nothing going back to the cutter adjustment module either.

I don't have SAB schematics here and don't know them as well. I'm curious though. I'll ask the guru.