Hello chaps,
Over the past few years I've designed and built a fair few esoteric phono preamps for older mono formats. During these endeavors I have discovered that for a practical preamp, the following features are absolutely essential:
U1.1 is configured as a standard 'all-in-one' amplify and equalise type stage, with provision for variable time constants via switchable capacitors using a pair of 6 pole rotary switches for the bass and treble time constants. A total of 36 curve combinations are possible, including the RIAA curve, of course. Flat gain is 37dB, as the 40dB I usually design for is pushing things a bit for 78rpm levels.
U1.2 performs the bandwidth filtering, using a pair of 3rd order Sallen-Key type filters. As the cut-off frequencies of the rumble filter, a static 25Hz, and the HF cut filter are so far apart, they can be included in a single amplification stage. By switching the 3 capacitors in the HF cut filter using a rotary switch, 4 different HF cuts can be obtained. Very useful. The output is then de-coupled with C7, R16 and R17 to give a low impedance line output.
So there you have it
.
I have included the list of equalisation turn-overs, roll-offs and their corresponding capacitor values in a PDF document attached to this thread. Also included, are the HF cut values.
As an example to demonstrate the necessity of a bandwidth limiting filter, I have attached (in a ZIP folder) 2 audio examples of Bruno Walter conducting the LSO, performing the overture from Mozart' 'Marriage of Figaro', recorded in 1932. Quite a noisy 78, very crackly. The limiting frequencies are 25Hz and 6.5kHz.
Over the past few years I've designed and built a fair few esoteric phono preamps for older mono formats. During these endeavors I have discovered that for a practical preamp, the following features are absolutely essential:
- Separate bass and treble turn-over and roll-off selection for the adjustable playback compensation. This allows a much greater combination of curves to be approximated, as opposed to 12 or so preset curves.
- Monaural operation, and an input stage to match. Parallel inputs from a stereo cartridge require equivalent parallel loading.
- A rumble filter. Ideally between 20Hz and 30Hz. I have yet to encounter anything remotely related to musical material below 35Hz on any transfers that I have made that were released pre-RIAA.
- For 78s - a 3rd order HF cut filter. Pre-war 78s exhibit an upper limit of 8kHz, usually 6kHz, with few exceptions. Post war 78s, possibly 12kHz on a good day. Acoustics, 3kHz. Early LPs may also be limited to 10kHz or so, to protect the cutter head from overheating. It is therefore useful to filter any noise above this limit, with a steep filter, especially in the case of 78s, where noise is triangular and can be very obtrusive (see attached examples, the difference is night and day).
- Excellent overload margins, with it's heavily modulated grooves and high linear velocity a 78 can give a preamplifier an awfully hard kick. Magnetic cartridges are velocity to voltage devices after all .
U1.1 is configured as a standard 'all-in-one' amplify and equalise type stage, with provision for variable time constants via switchable capacitors using a pair of 6 pole rotary switches for the bass and treble time constants. A total of 36 curve combinations are possible, including the RIAA curve, of course. Flat gain is 37dB, as the 40dB I usually design for is pushing things a bit for 78rpm levels.
U1.2 performs the bandwidth filtering, using a pair of 3rd order Sallen-Key type filters. As the cut-off frequencies of the rumble filter, a static 25Hz, and the HF cut filter are so far apart, they can be included in a single amplification stage. By switching the 3 capacitors in the HF cut filter using a rotary switch, 4 different HF cuts can be obtained. Very useful. The output is then de-coupled with C7, R16 and R17 to give a low impedance line output.
So there you have it
.I have included the list of equalisation turn-overs, roll-offs and their corresponding capacitor values in a PDF document attached to this thread. Also included, are the HF cut values.
As an example to demonstrate the necessity of a bandwidth limiting filter, I have attached (in a ZIP folder) 2 audio examples of Bruno Walter conducting the LSO, performing the overture from Mozart' 'Marriage of Figaro', recorded in 1932. Quite a noisy 78, very crackly. The limiting frequencies are 25Hz and 6.5kHz.
Attachments
Last edited:
It's been 24 hours.
Noted, schematic and details saved. Details about preemphasis and frequency limitations of 78's are valuable information I've never seen before.
Currently my 78 playing platform is a Howard player/radio combination my parents bought in 1949. The steel arm and needle are rather hard on the shellac.
Fortunately most of the 78 tracks I fancy were reissued by cheap labels on LP in the late 50's. My parent's postwar big band albums were destroyed in a 1956 move.
Enjoy your original opera cuts. Singing styles change; great performers who were written about cannot be heard except on orginal format. Unless someone like you saves them.
Noted, schematic and details saved. Details about preemphasis and frequency limitations of 78's are valuable information I've never seen before.
Currently my 78 playing platform is a Howard player/radio combination my parents bought in 1949. The steel arm and needle are rather hard on the shellac.
Fortunately most of the 78 tracks I fancy were reissued by cheap labels on LP in the late 50's. My parent's postwar big band albums were destroyed in a 1956 move.
Enjoy your original opera cuts. Singing styles change; great performers who were written about cannot be heard except on orginal format. Unless someone like you saves them.
Last edited:
There's not much to it, perhaps that's why there haven't been many replies yet .
I'd recommend getting a 78rpm turntable like a Garrard SP25, a Stanton 500 with a 2.8mil truncated conical stylus (from Expert Stylus), and building this preamp. The perfect 78rpm playback setup !
In case you are interested, bandwidth can be determined quite easily by looking at spectrograms. See attached.
.I'd recommend getting a 78rpm turntable like a Garrard SP25, a Stanton 500 with a 2.8mil truncated conical stylus (from Expert Stylus), and building this preamp. The perfect 78rpm playback setup
!In case you are interested, bandwidth can be determined quite easily by looking at spectrograms. See attached.
Attachments
I thought I'd list out the turn-overs, roll-offs and cut-offs that this preamp can provide.
Bass turn-overs
So for example, should you be listening to a circa 1940 recording made in the USA, then you would use 500Hz - 2.2kHz with a 10kHz treble cut filter.
Bass turn-overs
- Flat
- 250Hz
- 300Hz
- 400Hz
- 500Hz
- 750Hz
- Flat
- 1.6kHz
- 2.2kHz
- 3.18kHz
- 5kHz
- 7kHz
- 15kHz
- 10kHz
- 6.8kHz
- 4.7kHz
So for example, should you be listening to a circa 1940 recording made in the USA, then you would use 500Hz - 2.2kHz with a 10kHz treble cut filter.
Last edited:
Well, there are some mono magnetic cartridges out there. I think Grado does one for 78s. In that case you would want to make R1 and R2 100k to give an input impedance of 50k.
As most cartridges are designed for 47k loading, and will be connected in parallel in a mono unit such as this (mono gives lower noise and distortion when playing back mono recordings with a stereo cartridge), an equivalent parallel loading will be required, so 2 47k resistors in parallel, R1 and R2 as you see in this preamp.
Ceramic cartridges would require a completely different pre-amplifier, as they are displacement to voltage devices, not velocity to voltage devices.
As most cartridges are designed for 47k loading, and will be connected in parallel in a mono unit such as this (mono gives lower noise and distortion when playing back mono recordings with a stereo cartridge), an equivalent parallel loading will be required, so 2 47k resistors in parallel, R1 and R2 as you see in this preamp.
Ceramic cartridges would require a completely different pre-amplifier, as they are displacement to voltage devices, not velocity to voltage devices.
OK, but usually, the most changes are made in the feedback loop, in this case.
I made one more or less like yours, but I used automatic mag/cer change, using a simple HCF4066. In the cartridge, you can see that usually there are two live wires, and two grounded wires, the return. In the ceramic, both ground wires are shorted, and in the mag, not. Then, a resistor giving some filtered DC between the two (now AC) grounded, and a simple CMOS gate detects the short or not, and then, the 4066 makes the constants change needed in the feedback loop to play both cartridges without operator intervention.
If you like the idea, please give some credits $$$$ ;-D as it is not patented.
Quite a clever solution in your case.
I suppose you could stick an analogue switch at the bottom of the feedback loop so as to turn the gain-stage into a follower.
I don't really see the point in using ceramic cartridges any more, they have a low compliance which increases wear on the tip and the record itself. In this circuit, one would use a dedicated turntable for 78s with a single magnetic cartridge. Ideally a Stanton 500 with a 2.8 mil TC tip.
I suppose you could stick an analogue switch at the bottom of the feedback loop so as to turn the gain-stage into a follower.
I don't really see the point in using ceramic cartridges any more, they have a low compliance which increases wear on the tip and the record itself. In this circuit, one would use a dedicated turntable for 78s with a single magnetic cartridge. Ideally a Stanton 500 with a 2.8 mil TC tip.
Steel needles (or cactus thorns) cause some pretty nasty damage!
Have a listen to some of the transfers on my site. These were made with a modern cartridge. You can track a 78 at 5 grams or so quite easily. I've done a fair few experiments with steel needles and early pick-ups but I always find a Stanton 500 with a proper tip gives the best results.
Have a listen to some of the transfers on my site. These were made with a modern cartridge. You can track a 78 at 5 grams or so quite easily. I've done a fair few experiments with steel needles and early pick-ups but I always find a Stanton 500 with a proper tip gives the best results.
OK, I promise go to this site to listen some there.
I usually prefer to listen music in its original format. For example, in one of my 486's I have 1726 hours of house, tribal, hard, trance and other electronic music kinds. In the other hand, I strongly dislike CD's, I have a double album of Ana Belén and Victor Manuel (Two Spanish singer songwritters) called "Mucho más que dos" that has no external damage, but the disk is unusable (If you see though them, you will see tenths of minute holes in the aluminium layer). So I went to the BMG sales office, and say them that the disk is failed. A guy here gave me a copy, and immediately I copied to MP3, but is one of the few cases.
But listen, for example, a good vinyl in MP3 is a waste. I prefer to listen the original version.
I usually prefer to listen music in its original format. For example, in one of my 486's I have 1726 hours of house, tribal, hard, trance and other electronic music kinds. In the other hand, I strongly dislike CD's, I have a double album of Ana Belén and Victor Manuel (Two Spanish singer songwritters) called "Mucho más que dos" that has no external damage, but the disk is unusable (If you see though them, you will see tenths of minute holes in the aluminium layer). So I went to the BMG sales office, and say them that the disk is failed. A guy here gave me a copy, and immediately I copied to MP3, but is one of the few cases.
But listen, for example, a good vinyl in MP3 is a waste. I prefer to listen the original version.
Hmm... We can do a lot to reduce noise on older releases when they are in the digital domain. Some early CDs have rot problems, but I have only ever come across a couple.
MP3 is a bit rubbish, I only use lossless (FLAC). But if I have a good LP with few plays, then I always transfer it during the first time that I play it, lest it deteriorate.
MP3 is a bit rubbish, I only use lossless (FLAC). But if I have a good LP with few plays, then I always transfer it during the first time that I play it, lest it deteriorate.
In the UK, most of the 78s are very noisy, due to the weather . High humidity, cold temperatures etc. cause the shellac to deteriorate over time, with many houses not having central heating until 20 years ago.
This makes it much more challenging to repair noisy sections in the digital domain, although I am getting rather good at it now.
. High humidity, cold temperatures etc. cause the shellac to deteriorate over time, with many houses not having central heating until 20 years ago.This makes it much more challenging to repair noisy sections in the digital domain, although I am getting rather good at it now.
What 78's show up in Indiana are such dogs I probably won't buy some modern turntable & cartridge. Typical finds are no-name sweet dance bands in a Sammy Kaye album cover. I didn't even like Mr. Kaye.
OTOH I was listening to Caruso on LP reissue of "neopolitan songs" last night: I found at Goodwill for $1. No great melodies, Jerome Kern hadn't lived yet. I've got the Carter family in CD reissue, and Jimmy Rogers on LP, other 1920's acts I find interesting. Verdi & Puchini excerpts might be interesting but not likely to be found in this lower middle class state. There are plenty of opera excerpt covers out there on LP.
I'm not listening to digital music now "the shockwave plugin has been hacked do you really want to connect to that site?". and give away my brokerage passwords. When shockwave was working most of tracks on you-tube and soundcloud had pretty vile production values. Plenty of work to do backing up files & installing a newer linux op system on a new used computer my brother gave me so I can update past lubuntu11.
Have fun in the center of culture preservation, the UK, and cold clammy weather.
OTOH I was listening to Caruso on LP reissue of "neopolitan songs" last night: I found at Goodwill for $1. No great melodies, Jerome Kern hadn't lived yet. I've got the Carter family in CD reissue, and Jimmy Rogers on LP, other 1920's acts I find interesting. Verdi & Puchini excerpts might be interesting but not likely to be found in this lower middle class state. There are plenty of opera excerpt covers out there on LP.
I'm not listening to digital music now "the shockwave plugin has been hacked do you really want to connect to that site?". and give away my brokerage passwords. When shockwave was working most of tracks on you-tube and soundcloud had pretty vile production values. Plenty of work to do backing up files & installing a newer linux op system on a new used computer my brother gave me so I can update past lubuntu11.
Have fun in the center of culture preservation, the UK, and cold clammy weather.
I might even try and import some US pressings, as they are much quieter . Given what I can already do with British pressings, I reckon I could outdo all of the major re-mastering houses in the US given the same raw transfers as them 
.
I just use HTML5 audio. Much easier! Flash is a complete pain.
. Given what I can already do with British pressings, I reckon I could outdo all of the major re-mastering houses in the US given the same raw transfers as them .I just use HTML5 audio. Much easier! Flash is a complete pain.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.