Phantom power with no blocking caps in transformerless circuit?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Joined
Aug 7, 2009
Messages
35
Location
Oklahoma City, Oklahoma
There are no phantom blocking caps... I am curious how this works.

This a commercial product and it indeed works fine and I have visually verified there are no added caps on board for this purpose. What gives?

I suspect  it has to do with the additional circuitry between the +55 in and the gain control. It has a traditional phantom supply voltage developed elsewhere from +75.

Any ideas?

I tried to attach a simplified pic but the folder is full only brings an error.

Here is the link to the schematic.

https://ia800201.us.archive.org/23/items/symetrix_421M_1C00_sch/421M_1C00_sch.pdf

Regards
 
Constant current source from +55V. Input pair returns to -13V. I'd say the bases of Q12 Q13 can be put anywhere from +49V to -9V and they do not care.

I agree it could be drawn clearer; and probably was, on the original napkin.
 
MediaWorksOKC said:
There are no phantom blocking caps... I am curious how this works.

This a commercial product and it indeed works fine and I have visually verified there are no added caps on board for this purpose. What gives?
The input transistors are allowed to float up tens of volts... the collector currents are ground referenced by the op amps... Since the negative feedback doesn't wrap around the input transistors it isn't very low distortion...
I suspect  it has to do with the additional circuitry between the +55 in and the gain control. It has a traditional phantom supply voltage developed elsewhere from +75.

Any ideas?
what's the question?

JR
I tried to attach a simplified pic but the folder is full only brings an error.

Here is the link to the schematic.

https://ia800201.us.archive.org/23/items/symetrix_421M_1C00_sch/421M_1C00_sch.pdf

Regards
 
JohnRoberts said:
Since the negative feedback doesn't wrap around the input transistors it isn't very low distortion
Agreed. One may wonder if the benefit (?) of not having coupling capacitors is not largely outweighed by the fact that the differential pair is not included in the feedback circuit. Seems to me like answering a question not asked and creating another.
 
abbey road d enfer said:
...One may wonder if the benefit (?) of not having coupling capacitors is not largely outweighed by the fact that the differential pair is not included in the feedback circuit.

What's the real difference (absolute and relative) in distortion between a preamp where the NFB is used and ones where isn't?
 
moamps said:
What's the real difference (absolute and relative) in distortion between a preamp where the NFB is used and ones where isn't?

It depends on the amount of feedback (the difference between the open loop gain and the closed loop gain). In op amps this can easily be 60dB or more so the distortion will be reduced  by the same amount.

Cheers

Ian
 
ruffrecords said:
It depends on the amount of feedback (the difference between the open loop gain and the closed loop gain). In op amps this can easily be 60dB or more so the distortion will be reduced  by the same amount.

I'm familiar with feedback theory. I'm asking for this particular example where input diff. pair is used in front of an opamp. 
 
Sorry but didn't ask why the diff pair is used at all here.
JR and Abby said the preamp without global NFB is inferior. I just ask for which amount at what absolute amount. 
 
moamps said:
What's the real difference (absolute and relative) in distortion between a preamp where the NFB is used and ones where isn't?
The difference is about one order of magnitude (20dB), which can be quite significant at higher gain. Typically 0.2% @50dB gain 0dBu output for this circuit. With NFB applied to the emitters less than 0.02%. Actually it depends on how NFB is applied. Just using a Szlikai pair results in 20 dB improvement. Taking NFB from the opamp outputs (each transistor set must have its own opamp) results in significant additional improvement.
Some would say it's not significant because they can't hear it, but as i am unrepentently addicted to purity I think it is worth considering.
 
Thanks. I'm asking that because, for example, Soundcraft, MaCkie, Studer etc. used both topology (global NFB and Sziklai pair NNFB). There are obviously some other parameters except THD which may be count in.

So, could we do a modification of this DC coupled  preamp by using Sziklai pair and get a preamp which can be almost equal to a GNFB based preamp THD wise?
 
moamps said:
Thanks. I'm asking that because, for example, Soundcraft, MaCkie, Studer etc. used both topology (global NFB and Sziklai pair NNFB).
I have studied this subject with interest when Soundcraft introdudes the use of the Sziklai pair (in the Folio range IIRC). Indeed it was a step forward; the next was when they used global NFB (I just don't remember in which model it was introduced).

There are obviously some other parameters except THD which may be count in.
Indeed there is considerable improvement in common mode handling.

So, could we do a modification of this DC coupled  preamp by using Sziklai pair and get a preamp which can be almost equal to a GNFB based preamp THD wise?
I believe so. The thing is that unit was not intended at high-performance audio, rather typically at AM broadcast and paging. I don't know if Robert Orban was aware of the possible improvements or not willing...?
The subject of capacitorless mic preamps has been beaten to death
Check https://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=14&p=22&hilit=capacitorless+preamp&sid=90f3a970455a01fcb948a2977e610117#p22
Take your time, there are 300+ posts in this thread.  :)
JR is very active in this group.
 
moamps said:
What's the real difference (absolute and relative) in distortion between a preamp where the NFB is used and ones where isn't?
The difference absolute and relative is lower distortion, The Vbe voltage changes with emitter current, in operation this Vbe change Introduces distortion. Alternately connecting NF back to thin input pair operates the transistors constant current and eliminates that source of distortion.

Apparently passable preamps have been designed operating the input pair open loop, I vaguely recall one popular brit mixer used this (not the capacitor-less variant, but the lack of full NF). 

JR
 
We seem to be debating two different circuit aspects, eliminating DC blocking, and managing input pair linearity.

I am with Abbey about lower distortion being better, especially when the cost is insignificant.

Eliminating the DC blocking caps is a whole nutter topic.... For TMI I actually joined this forum to follow Wayne's capacitor-less effort (before he left in a huff, not because of me  8)  ). 

Here is 13 forum pages of his current musings on the subject  https://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=598.  In the decades before discovering Wayne's effort I had done a lot of my own circuit design scribbling after beer o'clock, and came up with multiple approaches (all mine used NF around the input pair). I never melted solder to make one and remain unconvinced that it is a merchantable mass market feature.

As Abbey said, it answers a question that nobody is asking.  ::) (Well very few are asking). In my judgement this is more of an esoteric marketing hook than real demonstrable (measurable ) benefit.

FWIW we could add a couple more poles to the phantom power switch and shunt across these blocking caps when phantom is not being used, but it would click and thump and be a pain to make audiophool friendly (they don't like clicks/thumps). 

JR
 
abbey road d enfer said:
I didn't know that. Wasn't he the group's  owner?
no AFAIK but he started hosting his own forum  https://www.proaudiodesignforum.com/forum/php/viewforum.php?f=1&sid=bc4a830a4f1f7bc807f9242897e60547
I'll have a look at it.
I have probably discussed it here too in passing over the years.
=====
I had a friend still designing high performance analog consoles (now RIP) and I could never get him to take it seriously, even as an esoteric feature.

JR 
 
JohnRoberts said:
no AFAIK but he started hosting his own forum  https://www.proaudiodesignforum.com/forum/php/viewforum.php?f=1&sid=bc4a830a4f1f7bc807f9242897e60547I have probably discussed it here too in passing over the years.
Actually I'm a subscriber there, but my last visit before today was about 18 months ago...
I had joined the former group because of the MicMix thread (not because I want to build one, but simply out of curiosity) and transferred to the new group, but it had slipped my mind...
I must say, as much as I respect Wayne for his endeavours,  I'm not that interested in capacitorless mic pres or phono pres.  :)
I like to keep informed, so I don't look like an idiot when someone asks about it.

I had a friend still designing high performance analog consoles (now RIP) and I could never get him to take it seriously, even as an esoteric feature.
It's about 20 years I haven't been involved in analog mixers but if I was, I would be reluctant to add a 72V rail, unless I could brag about increased headroom, like the SPL guys; 40Vrms output, that eventually ends up in a 5V converter chip. I would be a tad more motivated for current mixing (an idea several of us had, unfortunately but wisely rejected by THAT) and time-precedence panpot.
 
If you want the best possible performance and you're not mass producing cheap mixers, then just spend a few extra dollars and use a transformer. The best pre I have (in the purest interpretation of performance) is a THAT1512 (CFIA) circuit with a transformer in front of it with a center tap for 48V. A mic input transformer does a lot, eliminates a lot of parts and there are pretty good ones that are not terribly expensive.

Also, the effect of wrapping the transistors with op amps on THD is plotted quite exactly on p 55 of the PDF posted by user 37518.
 
squarewave said:
If you want the best possible performance and you're not mass producing cheap mixers, then just spend a few extra dollars and use a transformer.
Best?  I stopped using transformers in consoles back in the 70s... Good mic transformers are not cheap and low noise bipolar transistors could spank them on the test bench  by the late 70s.
The best pre I have (in the purest interpretation of performance)
I do not recognize that as an objective metric? Of course use whatever floats your boat.
is a THAT1512 (CFIA) circuit with a transformer in front of it with a center tap for 48V. A mic input transformer does a lot, eliminates a lot of parts and there are pretty good ones that are not terribly expensive.
Transformers have advantages in high RF environments and are somewhat more tolerant of poor facility grounding practices.

When I was designing inexpensive fixed install background music systems we used (cheap) transformers on input and output, not for fidelity but because the customer is always right, even when wrong.  ::)
Also, the effect of wrapping the transistors with op amps on THD is plotted quite exactly on p 55 of the PDF posted by user 37518.
I sure wish those experts would have written those books several decades earlier...  8)

JR 
 
JohnRoberts said:
Best?  I stopped using transformers in consoles back in the 70s... Good mic transformers are not cheap and low noise bipolar transistors could spank them on the test bench  by the late 70s. I do not recognize that as an objective metric?
Please enlighten me JR. Like I said, I qualified my post with the clause that we're not talking about cost. Otherwise, my understanding has always been that (cost aside) a transformer is superior in every metric. Even if noise of a high-perf transistor is less than a transformer under whatever confluence of circumstances you're citing, it seems to me you could use a 1:1. But another benefit of the transformer is that you can match the OSI of the transistors behind it. So if my understanding is incorrect, please explain the details under which an active circuit can beat a transformer in any metric, noise, cmrr, whatever. A transformer is a passive device. So where is this noise coming from that is being spanked?
 
Back
Top