Man, we're really getting some quotes going here!
[quote author="JLM Audio"]Hope this make some sense as I haven't had the greatest amount of sleep since my baby boy Ryland was born a couple of weeks ago. But I am not complaining because he is cool
So I hope this helps make more sense and not just confuse the issues.
[/quote]
congratulations!
(emrr)I'm curious why so many app notes go on about the importance of 1% matched resistors for balanced mic pads. Certainly they are not thinking about center tapped transformers, since they went out of fashion long before the common use of U pads for mics.
(JLM Audio)A pad would usually be separate device that was used with lots of different gear so should be precision matched so it works with lots of gear the same and since a lot of old gear did have there centre tap tied to ground it would be important. But in the controlled situation in a rack with a floating transformer as shown above it would not matter as the whole Common mode reduction is due to the floating primary coil. Putting any series resistance in series with any leg of the primary does not unbalance the CMR as there is no complete circuit path for the CMR. In simple terms is would be like a series circuit with battery and lamp and switch with the switch open and therefore no circuit to light the lamp. Ground the centre tap of the transformer primary and it is a completely different story as there would be a circuit and matched resistances in each leg would be needed.
Fair enough, if a single leg pad works right it works right.
(JLM Audio) The pad can attually be made with only a single pot in one leg of the balanced line like lots of vintage gear as matching the resistors will do nothing as shown below. We only use a dual pot as it makes the wiring easier.
I fail again to see how the dual pot makes the wiring easier than the single; maybe it?s just that a 2K pot is significantly harder to come by? Even the Jensen website still posts the precision matched info I'm referring to in regards to non-center-tapped transformers and onboard pads. I guess almost no one gets it, since balanced and tightly matched ?20db U pads are what shows up in 99% of products and are the conventional wisdom that almost always gets promoted. Heck, it?s what you use in the Go Between kits.
(emrr)if we put a true 150 ohm mic on this pad, with it set at the halfway point, we get (2 x) 1150 ohms in parallel at the wipers, or 575 ohms loading the transformer primary (415 ohms reflected to the mic, assuming a true 1500 ohms from the transformer(?)) that wants to see 150-200 ohms for proper Fi.
(JLM Audio)With the pad at halfway (linear) the load to the mic would be 500ohm + 500ohm + 1/(1/1150 + 1/1500) = 1651ohm. At full 20db attenuation load to the mic would be 1k + 1k + 1/(1/150 + 1/1500) = 2136ohms. At 0dB attenuation load to the mic would be 1/(1/2150 + 1/1500) = 884ohms. So I am not sure how you get 415ohms as the mic never sees a load lower than 884ohm. Also a 150ohm mic only ever wants to see 600ohms or greater load wise. Like a U87 for example could be strapped as 50ohm for 150ohm to 300ohm load mic pres or 200ohm for 600ohm to 1200ohm mic pres which is why the 1073 has the input impedance switch on the back panel. Most modern mic pres load these mics with 2 to 3k load. If the 150ohm mic had of been plugged direct into the 1500ohm transformer without the pad it would have had a 1500ohm load on it not 150ohm.
Yes 415 is incorrect for sure; I?ve doubled up on a value. I?m not arguing about the bridging load that the mic should see.
I?m more concerned with the load presented to the transformer by the mic and pad elements, which appears to have lower load resistances at the no pad and the full pad positions, with highest transformer load resistances at the center of the pot sweep. In my mid-point example, I?m considering the square as completed by the mic, the (2) 1K pots, and the 150 ohm resistor as appearing the same as two paralleled 1150 ohm resistors to the transformer, or 575 ohms. Which may be wildly incorrect.
More relevance to consider below the next set of quotes....
(emrr)The whole point of the U pad made with fixed resistors seems to be maintenance of the 150 load on the transformer primary, and the 1200-1500ish bridging load on the mic.
(JLM Audio)Actually the opposite is the case is you want to keep mic tone the same. If the mic is feeding a 1500ohm input and you switch a pad in that gives the mic a 1500ohm load the tone will stay similar. In 80's neve desks the the load to the mic would be low 1.2k with the pad out and about 13k with the pad in and the tone difference on dynamics and transformer output condensers can easily be heard. The transformer primary doesn't care in simplified terms about what resistor is has across it while the pad is in as reflections are minimum as this is a bridging system and not a matched load system. 600ohm to 600ohm matched load systems do want the mic and transformer to be loaded correctly to maintain the matched load.
I think you are reading me backwards, or we?re really saying the same thing looking at the load presented to the mic. So, I?m going to spout tangential info about the issues I?m wondering about with the load presented to the preamp by various states of the pad (or even types of pads), and how and when it is more or less critical.
Re-reading my quote above, a standard U pad load to the mic stays roughly the same as what the transformer presented. The standard 150-200 ohm shunt resistor replaces the mic load on the transformer, since the mic is now on the end of a pair of 600-700 ohm build out resistors, which to my understanding then would present the transformer with improper loading to achieve correct frequency response, if the 150-200 ohm shunt resistor is not present and there is no secondary load on the preamp transformer. Which is beside the point, since I?m pretty sure the build-out alone won?t reduce any significant mic voltage without the shunt present, it will only change the load to the mic and the transformer in undesirable ways. Maybe a variable source load to the preamp transformer is not an issue with some modern iron, but it could be an issue to folks re-furbishing or building with vintage iron, which in many cases wants to see the specified 150-250 ohm source load for full frequency response. Which is not to say that the vintage iron is true 150 ohm input, it?s just specified to work with that source. The opposite side of the spectrum being that a load on the transformer which is smaller than the specified 150 ohm source will extend the frequency response of the transformer to a degree.
I need to run some load tests with various vintage iron and see what impedances are really being presented to some test mics, since it?s virtually impossible to compare the way old iron is specified and new iron is specified with regards to source loading. It?s simply not addressed with old catalog information, and antique circuits are almost always run with unloaded secondaries. Which is what most people will run into if restoring or cloning old classics.
And as I suggested, maybe you are making the assumption here that one will always have a secondary load network of some type, thus rendering my questions moot. And maybe there can be a lot more load variation presented to the preamp input than I currently accept. Could be useful info for others to consider as well.
(JLM Audio)
I spent 16 years designing and working with bridged and matched balanced systems in Telstra which were usually twisted pair unshielded several kilometres long and was always amazed that we could get broadcast quality audio out the other end. :grin:
now that's a party! Sorry to be so nitpicky; trying to understand the nuts and bolts of it better.
