just done some digging and the 80B series did indeed use TL07x. I like the quote on the Trident webside of the model A
Some very memorable tracks recorded on it though. Trident Studios - Wikipedia
Some very memorable tracks recorded on it though. Trident Studios - Wikipedia
That's what i heard.
Can't confirm since so far nobody let me open up their Trident to have a look.
Hmmm....interesting. They MUST have used discrete components for their mic preamps, though--methinks the TL07x would be far too noisy for that application.
60dB gain would be totally impractical from a transformer yes, the impedance ratio would be 1,000,000 : 1. Meaning every pF of parasitic capacitance on the secondary becomes 1uF when referred to the primary.
Fortunately to have a TL071 perform noise-wise like a low noise bipolar probably needs no more than 1:20 step up.
Fortunately to have a TL071 perform noise-wise like a low noise bipolar probably needs no more than 1:20 step up.
The main purpose of the transformer is not to get 60dB of gain, but to present the optimum source impedance to the opamp while raising the gain a bit, thus keeping the noise figure low, the transformers used are step up transformers (as opposed to step down like the one you mentioned), the ratio is dependent on the equivalent noise voltage and current of the opamp. FET opamps like the TL07x have very low noise current, thus the required optimum source impedance is quite high, requiring higher ratio transformers, fortunately it turns out that NF wise, usually something like a 1:10 transformer will suffice to keep the noise figure sufficiently low on a FET opamp. Considering a 150 ohm mic impedance, in the case of a TL07x this means a NF of around 3dB using a 1:10 transformer, as opposed to the 21dB NF you get by using no transformer, a higher ratio transformer would lower the NF even more, but as the ratio of the transformer departs from 1:1 it tends to become less than ideal, for instance Jensen transformers mic transformer line dont go beyond 1:10. Bipolar opamps usually require mid to low ratio step up transformers.
Last edited by a moderator:
My favourite bad op amp made good idea was a 741 with 2 external transistors in long tail pair with perhaps a 10K pull down resistor to the output ( or pull up if prefered ). As John said he saw it often in the Pro world before the NE5534 came along. John I seem to remember has 130 microphones. In his opinion the boosted 741 is still good enough to use his best microphones. That's not to say the That's op amp he uses now isn't better.
Here is something that offers low noise in the spirit of the 741 idea. If you look at the NE5534 internal circuit you will find the compensation pins go directly to suitable points for an external LTP. The inputs (2,3) are biased off. That should mean specification is maintained, noise should be better. I like the idea of BD139. Without ever having tested it I would think BC337-40 could be good. John said at SSL they thought that under ideal conditions -90dB could be possible from a microphone ( I strongly doubt it, however I know very little of state of the art devices or theoretical Boltzmann's ( Johnson's or whatever ) noise levels of an unhindered ribbon, 0R1= 6nV? ). He says it degrades to about -76db in the real world.
Here is something that offers low noise in the spirit of the 741 idea. If you look at the NE5534 internal circuit you will find the compensation pins go directly to suitable points for an external LTP. The inputs (2,3) are biased off. That should mean specification is maintained, noise should be better. I like the idea of BD139. Without ever having tested it I would think BC337-40 could be good. John said at SSL they thought that under ideal conditions -90dB could be possible from a microphone ( I strongly doubt it, however I know very little of state of the art devices or theoretical Boltzmann's ( Johnson's or whatever ) noise levels of an unhindered ribbon, 0R1= 6nV? ). He says it degrades to about -76db in the real world.
My favourite bad op amp made good idea was a 741 with 2 external transistors in long tail pair with perhaps a 10K pull down resistor to the output ( or pull up if prefered ). As John said he saw it often in the Pro world before the NE5534 came along. John I seem to remember has 130 microphones. In his opinion the boosted 741 is still good enough to use his best microphones. That's not to say the That's op amp he uses now isn't better.
Here is something that offers low noise in the spirit of the 741 idea. If you look at the NE5534 internal circuit you will find the compensation pins go directly to suitable points for an external LTP. The inputs (2,3) are biased off. That should mean specification is maintained, noise should be better. I like the idea of BD139. Without ever having tested it I would think BC337-40 could be good. John said at SSL they thought that under ideal conditions -90dB could be possible from a microphone ( I strongly doubt it, however I know very little of state of the art devices or theoretical Boltzmann's ( Johnson's or whatever ) noise levels of an unhindered ribbon, 0R1= 6nV? ). He says it degrades to about -76db in the real world.
Thats a cool circuit, you can also do something similar with the LM301
A lot of engineers used the older Neve desks for recording. The 1081 input modules ( all discrete, transformers for all ins/outs) are still sought after.
neve 1081 - Google Search
Picture of the inside of one.
I own an MCI JH-500 console, it is packed with NE5534 and I absolutely love it.
Its fully differential, low noise and very low distortion. Unfortunately none of the latest generation opamps give access to the internal nodes allowing for similar tricks. The substantially increased GBW of the latest suggest some pretty incredible performance could be had if they were available.
Don't take me at gospel (ever, but especially now), but didn't JCX mention at some point that bypassing the diff pair didn't really matter as much with the high GBW of the newer chips? Obviously the compensation scheme changes up.
Scott Wurcer ages ago threw up some pretty clever diff pairs RIAA preamps (now my memory is getting REALLY FOGGY) that folded in one of the higher-current higher-speed opamps as both a loop gain booster and output buffer.
I haven't earnestly played around too much though, so I'm definitely speaking from second-hand experience.
Scott Wurcer ages ago threw up some pretty clever diff pairs RIAA preamps (now my memory is getting REALLY FOGGY) that folded in one of the higher-current higher-speed opamps as both a loop gain booster and output buffer.
I haven't earnestly played around too much though, so I'm definitely speaking from second-hand experience.
I'm told broken up Neve desk sections get built into modern desks so as to be able to get the sound. As far as I know Neve never made money. I believe the transformers were made in house. When I was looking to join the AES only Dr Sowter was willing and able to vote me on ( I think he and Ray Dolby were founding memebers ). Peter Walker didn't have his PHD then. I remember the 97 year old Sowter telling me about how with care a transformer could have very low distortion. Surprisingly British Telecom being a customer for this. One of his last lectures was on this subject. From the litlle I know we can get a better noise performance using a transfomer regardless of op amp used. Sowter was a wonderful person to talk to. As is usually the case he made the complex simple as he knew the key facts. You always know a person like this because like a car they stay in the gear they need when talking. If the road allows the higher gear is selected. Richard Feynman also. His simple frozen rubber experiment for whatever reason was too simple for NASA. It was the reason and complex reasons were wrong reasons.
I searched this thread to see if anyone actually listened to a Sparkos SS3602/1 but only found a few opinions. Glanced at data sheet, class A bias so I assume it’s not class (A)B output.
Two pole compensation mentioned too. I was looking at it for several reasons , through hole and since standard 8 pin pins it is easy to do comparisons.
Sort of looking more for user report data not opinions. Thanks
Discrete Op Amps - SparkoS Labs Inc.
-
Two pole compensation mentioned too. I was looking at it for several reasons , through hole and since standard 8 pin pins it is easy to do comparisons.
Sort of looking more for user report data not opinions. Thanks
Discrete Op Amps - SparkoS Labs Inc.
-
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.