RS127 RS135

[mjrippe]

I just simulated the first stage. -3dB point is 124Hz. SLope is a gentle 3dB/octave until the inductor kicks in around 20Hz and it increases to 6dB/octave. Definitely no presence boost.

Cheers

Ian

I am guessing that the Dialog EQ (first stage) would not have the boost, but that the Film Loss EQ (second stage) might.  I'm sure you caught the drawing error with all those parallel R/C parts.  Since there are ten wires coming from the rotary switch it only makes sense that they would be switched as well.  Also worth noting that the inductor label shows the values at 10KC.

Mike

 

[ruffrecords]

I am guessing that the Dialog EQ (first stage) would not have the boost, but that the Film Loss EQ (second stage) might.  I'm sure you caught the drawing error with all those parallel R/C parts.  Since there are ten wires coming from the rotary switch it only makes sense that they would be switched as well.  Also worth noting that the inductor label shows the values at 10KC.

Mike

The Film Loss EQ looks basically the same as the Dialog EQ but with different values. I agree the way the schematic is drawn is  odd. One switch is shown clearly as a switch but the others are not. The inductance values being quoted a 10KHz is important from the point of view of ensuring there are not spurious responses at high frequencies. All inductors have a self resonant frequency at which point they become relatively high resistance and in this  circuit that could cause a high frequency boost (I know this occurs because it happens in my Pultec EQ). . The fact the manufacturer states the inductor is still just an inductor at 10KHz implies the self resonant frequency is well above the audio band.

However, to make sure, I will also simulate the Film Loss EQ and report back.

Cheers

ian

 

[Lee_M]

am interested in seeing if they managed to avoid the inductor self resonance problems I am having with a couple of my EQs and if so how they did it.

Just out of interest, Which circuits and inductors are you having resonance problems with?

Is the self-resonance only present at certain taps of your inductors, or do you have a consistent resonance (on all taps) that changes frequency?

I'm only a rookie, So apologies if this is obvious and you've already looked into it, But I wonder if modern core materials with extremely high q could be causing the problem?

 

[ruffrecords]

Just out of interest, Which circuits and inductors are you having resonance problems with?

Is the self-resonance only present at certain taps of your inductors, or do you have a consistent resonance (on all taps) that changes frequency?

I'm only a rookie, So apologies if this is obvious and you've already looked into it, But I wonder if modern core materials with extremely high q could be causing the problem?

The self resonance only manifests itself on the lower taps of inductors of 1H and 2H total inductance; specifically the Carnhill VTB9042 and VTB9050. The self resonance, which occurs as a dip around 26KHz, happens mainly on my 3 band Pultec design and to a lesser extent on the Helios 69 design. The effect is the reduce the apparent peak frequency and steepen the slope above the desired resonance. This is only noticeable on the highest frequency settings of the EQ. As it happens outside the normal audio band, nobody had detected it until someone needed to measure the response up to 48KHz for their application.

The reason it happens because the highest inductance value needs a lot of turns and adds stray capacitance.  It does not occur in my REDDEQ design which uses a very similar circuit and has a tapped inductor, the VTB9044, with a maximum inductance of 200mH.

Cheers

Ian

 

[Michael Tibes]

...It does not occur in my REDDEQ design which uses a very similar circuit and has a tapped inductor...

I also have very little experience in developing with inductors, but Isn't that the solution? In the german Vxx modules most of the transformers and coils are tapped. I could imagine that EMI did the same? I guess there are not many options to reduce stray capacitance, either live with it or tap the coil?

Would it be of interest to check the german PF5/8h module - which is a passive 34dB loss presence boost unit? I only have a simplified schematic, but I could try to take a peek inside.

Michael

 

[ruffrecords]

I also have very little experience in developing with inductors, but Isn't that the solution? In the german Vxx modules most of the transformers and coils are tapped. I could imagine that EMI did the same? I guess there are not many options to reduce stray capacitance, either live with it or tap the coil?

Would it be of interest to check the german PF5/8h module - which is a passive 34dB loss presence boost unit? I only have a simplified schematic, but I could try to take a peek inside.

Michael

Sorry Michael, I did not make myself clear. The inductors I use in the Pultec and Helios EQs are tapped. The problem comes when the largest tap has a fairly large inductance; the additional capacitance this causes seems to adversely affect the self resonant frequency of the lowest inductance  (highest frequency) tap.

This is all tied up with the basic design factors involved in a peaking EQ. The main one is the circuit impedance. The classic EQs of old were designed for 600 ohm circuits. This means the required inductance for a given peaking frequency and  sharpness is quite small. Small inductors have fewer turns, less stray capacitance and therefore a much higher self resonant frequency. My original poor man's Pultec had no inductors and was designed for modern 10K impedance circuits so there was no self resonance problem. When I was later persuaded to add a peaking mid/boost, the inductor values required were therefore a lot higher than in the original  and this is the root cause of the self resonance problem. I later added an in peaking hi boost like the original and again this required much larger value inductors with the same self resonance problem.

Since the self resonance occurs well above 20KHz it is not a problem for most users. However, for future versions there are ways round the problem. The simplest is to use separate inductors for the small value inductances so they are not affected by stray capacitance of the larger taps but this does require a PCB revision. An alternative solution is to make a small add on PCB with discrete inductors on it that will replace the original inductor. This would provide an interim solution for existing boards and a simple PCB layout revision for a later version. There are a couple of problems with this approach. First you cannot fit in enough discrete inductors to equal the original inductor values which means the Q (sharpness) of the peaking EQ will be reduced. Second, the discrete inductors have much higher dc resistance compared to their tapped versions which again reduces the sharpness of the peak.

An alternative solution is to lower the circuit impedance which will require lower value inductances. If we were to reduce it from 10K to about 3K then we could reduce the inductances by the same amount and the self resonant frequencies should then be out of harm's way above 80KHz. Since most modern pro gear will drive a 600 ohm load this should no be a problem.

Cheers

ian

 

[joaquins]

Using separate inductors instead of series would allow to make the DCR lower, you probably can get away with just 2 inductors, one only for the lower ranges with a few taps to do so and another for the higher ranges, same thing, but without connecting them in series. I know the space is a constrain for this, but the lower value inductor would be quite smaller than the other, the other could just be the original one.

JS

 

[Lee_M]

This is all tied up with the basic design factors involved in a peaking EQ. The main one is the circuit impedance. The classic EQs of old were designed for 600 ohm circuits. This means the required inductance for a given peaking frequency and  sharpness is quite small. Small inductors have fewer turns, less stray capacitance and therefore a much higher self resonant frequency.

Wasn't the abbey road stuff designed for 200Ω outputs?
If so, The inductances they used would be even lower (for the most part) than for 600Ω circuits, plus any resonant frequencies due to stray C (of which there would be less to begin with, due to less windings) would be pushed even higher up into the frequency spectrum...Right?

I've been toying with a "passive presence box" design in my spare time for the past few months, based on my best guess (constant-z attenuators and LC filters) for how the RS127 does its thing.
All of the L values I've been getting from my calculations have been very low, The highest value tap I need is around the 43mH mark.

 

[Michael Tibes]

If I got things right the presence box was built to go between the mic and the preamp, so the impedances would have been quite low. The PF5/8h needs a generator impedance less than 50R and a load higher than 2k.

Lee, how far did you get with your presence box? Would you mind sharing?

Michael

 

[Lee_M]

Lee, how far did you get with your presence box? Would you mind sharing?

I haven't got around to drawing up a decent schematic yet, I just have a few pages of calculations, component values and other scribbles which I'm working from.

It's basically a bridged-t attenuator (unbalanced attenuator, with transformers for balancing the input and and output) with series and shunt LC components built into the attenuator arms.

The complex aspect is putting boost and cut (10db in either direction) on the same rotary switch, With "0dB" as the mid point.
As the switch changes over from "boost" to "cut", the LC components need to change from series-connected to parallel-connected, Along with changes to the L & C values...So that switch assembly is likely to become fairly complex!

 

[ruffrecords]

Wasn't the abbey road stuff designed for 200Ω outputs?
If so, The inductances they used would be even lower (for the most part) than for 600Ω circuits, plus any resonant frequencies due to stray C (of which there would be less to begin with, due to less windings) would be pushed even higher up into the frequency spectrum...Right?

I've been toying with a "passive presence box" design in my spare time for the past few months, based on my best guess (constant-z attenuators and LC filters) for how the RS127 does its thing.
All of the L values I've been getting from my calculations have been very low, The highest value tap I need is around the 43mH mark.

Yes, spot on. 200R circuits will have even smaller L values than 600R with consequent benefits.

Cheers

ian

 

[ruffrecords]

Using separate inductors instead of series would allow to make the DCR lower, you probably can get away with just 2 inductors, one only for the lower ranges with a few taps to do so and another for the higher ranges, same thing, but without connecting them in series. I know the space is a constrain for this, but the lower value inductor would be quite smaller than the other, the other could just be the original one.

JS

Separate multi-tap inductors is definitely the best technical solution but does need more PCB space. It would be nice to keep the CPB area about the same as that makes possible the 3U Pultec with built in tube gain make up PCB in a single Eurocard.

Cheers

Ian

 

[joaquins]

Separate multi-tap inductors is definitely the best technical solution but does need more PCB space. It would be nice to keep the CPB area about the same as that makes possible the 3U Pultec with built in tube gain make up PCB in a single Eurocard.

Cheers

Ian

  Tricky, then you need a multi tap inductor without the high capacitance in the lower ranges. Having separate windings would help as it would have capacitance from the small winding to the large, then to the large to the wold, instead of being always all the capacitance to the world. Switching may be trickier as you do need to disconnect the second part of the winding, a single rotary won't do the job anymore. I don't know what impact would the open winding would do, shouldn't be so bad but in a high impedance circuit an open winding could be calling for miss behavior.

  Thiner wire and more insulation wouldn't help I guess for the higher DCR. Shielding either as it doesn't reduce the capacitance, only gives the ability to redirect it.

JS

 

[Bo Deadly]

I have no idea what your circuit looks like but what if you bootstrap the ground pin. Meaning drive G on the VTB9050 with a low Z follower. Would that not reduce the effect of the capacitance?

 

[ruffrecords]

  Tricky, then you need a multi tap inductor without the high capacitance in the lower ranges. Having separate windings would help as it would have capacitance from the small winding to the large, then to the large to the wold, instead of being always all the capacitance to the world.
JS

I don't know if capacitance to the world is a problem.  My passive EQ designs, like most others, use series resonance circuits. At resonance the circuit is low impedance and at all others it is high. The low impedance at resonance is what produces the boost. The problem is the the inductor itself has a stray parallel capacitance across itself that forms a parallel resonant circuit at a high frequency (self resonance). If this is close enough to the desired resonance it steepens the high frequency end of the desired peak and then at even higher frequencies, as the stray parallel capacitor begins to dominate, it produces an unwanted shelving boost. As possible solution would therefore be to change to using a parallel resonant circuit.  That way the stray capacitance would just be in parallel with the main capacitor and slightly alter the peak position but not the overall shape. I don't know if the complicates the switching or not but I think it is worth looking at.

Cheers

Ian

 

[joaquins]

I don't know if capacitance to the world is a problem.  My passive EQ designs, like most others, use series resonance circuits. At resonance the circuit is low impedance and at all others it is high. The low impedance at resonance is what produces the boost. The problem is the the inductor itself has a stray parallel capacitance across itself that forms a parallel resonant circuit at a high frequency (self resonance). If this is close enough to the desired resonance it steepens the high frequency end of the desired peak and then at even higher frequencies, as the stray parallel capacitor begins to dominate, it produces an unwanted shelving boost. As possible solution would therefore be to change to using a parallel resonant circuit.  That way the stray capacitance would just be in parallel with the main capacitor and slightly alter the peak position but not the overall shape. I don't know if the complicates the switching or not but I think it is worth looking at.

Cheers

Ian

  Maybe splitting the inductor's windings helps, same original inductor but a break after the smaller windings and then continue with the bigger ones. In that case you do need a switch to connect the two sections together once you are using the higher ones. If you can live with even bigger capacitance for the higher ranges you could use alternate sections of the higher and lower windings and get even smaller capacitance for the lower ranges, that's harder to wind though, the other option just needs an extra tap and the switching.

JS

 

[ruffrecords]

  Maybe splitting the inductor's windings helps, same original inductor but a break after the smaller windings and then continue with the bigger ones. In that case you do need a switch to connect the two sections together once you are using the higher ones. If you can live with even bigger capacitance for the higher ranges you could use alternate sections of the higher and lower windings and get even smaller capacitance for the lower ranges, that's harder to wind though, the other option just needs an extra tap and the switching.

JS

There may be a hybrid solution - use a smaller large inductor and wire it in series with a separate smallest value inductor. The smaller large inductor should occupy less space and allow room for the additional smallezt value inductor.

Cheers

Ian

 

[Whoops]

Anybody have EMI RS127 or RS 135 brilliance module I could borrow for a few days to look under the hood and measure? I am interested in seeing if they managed to avoid the inductor self resonance problems I am having with a couple of my EQs and if so how they did it.

Cheers

Ian

Hi Ian, in a studio I work regularly there's 2x EMI TG12412 EQ,  I could do some measurements for you and take pictures of the insides in case that helps. Just let me know how can I help

I actually had the 3 brilliance modules in Abbey Road when I produced a record there at studio 3, never seen them in any studio outside Abbey road.

Don't know if the TG12412 High end EQ part is similar to the any of the brilliance modules, but it might be.
There's people with the TG12412 schematic around but I dont have it and can't seem to find it

 

[ruffrecords]

Hi Ian, is a studio I work regularly there's 2x EMI TG12412 EQ,  I could do some measurements for you and take pictures of the insides in case that helps. Just let me know how can I help

I actually had the 3 brilliance modules in Abbey Road when I produced a record there at studio 3, never seen them in any studio outside Abbey road.

Don't know if the TG12412 High end EQ part is similar to the any of the brilliance modules, but it might be.
There's people with the TG12412 schematic around but I dont have it and can't seem to find it

Thank you for the kind offer. I have seen the TG schematics ad the EQ is an active one, unlike the RS series which were passive.

Cheers

ian

 

[Whoops]

Thank you for the kind offer. I have seen the TG schematics ad the EQ is an active one, unlike the RS series which were passive.

Cheers

ian

I remembered also that they also have a Big passive EMI equalizer unit.

It is the same as this one but it's green and not white:

Let me know if you need pictures , measurements, etc...