Topic: DPA 4060's with MT 24/96

[gorlando]

I just received my brand new DPA 4060's and, I am trying to understand how to better use them with my MT 24/96.

I planned to connect them to my mic2496 and then go digital into the MT. I am waiting for the breakout cable from CoreSound (hope that Len acts quickly in sending it) to test this.

However, I am also trying to understand if I can connect the 4060's directly to the MT with reasonable results. This would be for “super-stealth” situations or when I don’t want to bother with too many boxes (the MT, the MIC2496 and the battery extender for the MT!!).
I have gone through the TS posts, extremely useful as usual, and recap what I understand as follows:
•   It is possible to feed the 4060 from the 1/8” mic input of the MT, however the 4060 signal is too high for this input.
•   The following logical step is to connect the 4060 to the ¼” TRS input. In this case I understand the signal is OK, but the power supplied by the MT could not be enough. In fact I red in a post (sorry I forgot by whom) the use of a special “dongle” combining the signal connection to the ¼” TRS and the power taken from the 1/8” input.
•   The use of a battery box is generally recommended, with a 9V alkaline battery feeding the 4060’s through a resistance, the value of which is not clear to me but can be in the range 2.2 to 15 kΩ (!?!). A condenser is then used to block the DC to the MT inputs. This condenser can be between 1 and 2.2 µF depending on the impedance value of the ¼” TRS MT inputs. This impedance is not published by M-Audio and therefore the good practice seems to be the “overkill” by using the 2.2 µF condenser.

In this scenario I decided to do something to understand better by myself and this is the summary of my main findings:
•   First, I posted a query on the Micbuilders Group on Yahoo (micbuilders@yahoogroups.com, which I strongly recommend for those interested in microphone DIY) asking for help on the battery box R and C values. For the 4060 – MT combination. A very useful reply, based on 4061’s was suggesting the overkill (2.2 µF) approach for the condenser, and a 15 kΩ resistor (for 9V battery). Apparently this combination was tested with very good results. I found a value of 10 kΩ mentioned on several TS posts, so I think 10 kΩ is probably a better choice. I understand that DPA recommends feeding the 4060’s with 5 to 50 Volts and with a current of 1 to 10 mA.

•   Then I decided to see what the MT was able to supply, by feeding one 4060 with temporary connections (crocodiles and the like). The results are:
o   1/8” input: voltage=2.52 V, current= 0.90 mA
o   ¼” TRS input: voltage=2.59 V, current=4.02 mA
In both cases I recorded (my voice) and the result appeared very good.
Then I connected both 4060’s to the ¼” TRS inputs: voltage=2.592 V, current=4.02 mA.
I also measured the voltage at one ¼” TRS input without mic connected, while the other input had a mic connected: the voltage was 29.92 V, which is in very good agreement with the 30 V provided when both inputs were “open”.

Some considerations, based on the Ohm’s Law:
•   The impedance of the 4060’s (calculated by the V, A values measured with a normal multimeter) is 645 Ω in the case of feed from the ¼” input, while it appears to be 2.8 kΩ with the 1/8” input.
I don’t understand this difference. This could be due to the 1/8” input having an additional impedance of 2.155 kΩ compared to the ¼” TRS inputs but, since the voltage at the 4060 terminals was 2.52 V this is inconsistent with the 645 Ω impedance! Anybody can help here?
•   Anyway the current provided by the 1/8” input (0.90 mA) is significantly lower than that provided by the ¼” inputs (4.02 mA). Since also the voltage provided is marginally higher (2.59 vs. 2.52 V) I think the ¼” TRS inputs are better than the 1/8” inputs not only for signal compatibility but also from a power supply point of view.
•   Considering the battery box (9V) alternative, as said before, with a resistor of 10 kΩ, the current provided to the mic is between 0.70 and 0.85 mA (considering the 4060 impedance being 645 Ω or 2.8 kΩ), with the voltage at the 4060’s terminals being between 1.97 and 0.54 V.
Lowering the resistor of the battery box to 2.2 kΩ, the above values become:
•   Current:  3 to 1.8 mA;
•   Voltage across the 4060: 2.04 to 5.04 V.

Based on the above I think that the MT can feed the 4060’s from its ¼” TRS inputs (stealth “on”). A battery box with a resistor of 10 kΩ seems to me not adequate, while a resistor of 2.2 kΩ raises its performances to the minimum targets in terms of voltage and current values as defined by DPA.
Since my target is to limit the number of boxes, I am going to test the MT feed first.

The above is posted to get comments / contributions and eventually to activate a discussion on the subject.
Giovanni

[guysonic]

While I'm not a fan of the DPA 4060 series because of inherent coloration and inaccurate pressure phase response (this discussion was posted late last year, but unfortunately removed from the archive by one the moderators), this mic seems to work best with 9 volt battery box using ~4.3 -to - 6.2K supply resistor.   Higher supply voltages will change best series resistor to be of a higher ohmage.  Determined this when evaluating these capsules when they first appeared ~1998 as I had hoped to use these for the DSM product line.

[gorlando]

I have done some further work to understand how to feed directly the microphones DPA4060 from the Microtracker (MT) 24/96 so to avoid a battery box and reduce the rig to simply the MT24/96, the DPA4060 and eventually a battery extender for the MT.

I started making some measurements of the characteristics of the MT 1/4" TRS inpur (phantom power on). The summary of the results is as follows:

Load, kΩ   R-S, V   T-S, V   R-S, mA   T-S, mA   R-S input   T-S input                                  R, kΩ    R, kΩ
    0   30   30   0   0   n.d.   n.d.
  45000        26.04        26.05        0.579        0.579        6843          6823
  30000        24.42        24.43        0.814        0.814        6855          6840
  15000        20.58        20.6          1.372        1.373        6866          6845

Therefore it appears that the 30 V phantom powers is passing through a 6.8K resistor (as per the phantom power spec) per each of the two balanced lines (Tip and Ring of the TRS). This was a good surprise to me, after reading tons of bad things about the weakness of the MT phantom power supply.

Comforted by these findings, I then utilized the schematic diagram provided by DPA for its adapter DAD6001, microDot to XLR (P48), designed to connect the DPA4060 to a 48 V phantom power supply.
I made the following considerations / calculations to adapt this schematic (attached) to the 30 V phantom power provided by the MT (instead of the standard 48 V).

Case 1
By analyzing the schematic diagram of the DPA DAD6001 it can be deducted that:
•   If the 4060 is fed with a tension of 8.2 Volts max. (fixed by the Zener diode), the relative calculated current is 0.82 mA (OHM's Law)
•   Considering the MT 24/96 (30V vs. 48V) feeding the same 0.82 mA to the load (which is 0.41 mA per each line of the balanced power supply (Tip and Ring), then the resistors total value (Rint+R2B+R1//R3) has to be 26.511 kΩ
•   Therefore each resistor R1, R2B and R3 for use with the MT 24/96 phantom power has the value of 15.407 kΩ (instead of 30 kΩ of the DAD6001).

Considerations
I was initially thinking to remove the 8.2 V Zener, but it is better keeping it to ensure the maximum voltage applied to the mic does not exceed 8.2 V. This Zener is never intervening in normal operation (since the Zener characteristic is non-linear and would cause distortion in the signal), so in normal operation the maximum voltage applied to the 4060 should be significantly lower than 8.2 Volts.

Case 2
The minimum supply voltage recommended by DPA is 5 Volts. If we consider this value in our calculations we have:
•   With the 4060 fed with a tension of  5.00 V (below the Zener threshold), the relative calculated current is 0.89 mA (OHM's Law)
•   Considering the MT 24/96 (30V vs. 48V) feeding the same 0.89 mA to the load )i.e. 0.44 mA per each line of the balanced power supply (Tip and Ring), then the resistors total value (Rint+R2B+R1//R3) has to be 28.140 kΩ
•   Therefore each resistor R1, R2B and R3 for use with the MT 24/96 phantom power has the value   of 16.493 kΩ (instead of 30 kΩ of the DAD6001).

Case 3
Following another hint, derived by the declared total consumption of the DAD6001 not exceeding 0.5 mA, we have the 4060 fed with a current of 0.5 mA. The relative calculated voltage drop in the resistors is then 24.20    Volts, which gives 23.80 Volts fed to the 4060 !! This is not possible, the Zener would intervene before. So this case is discarded.

Case 4
Another case descends from the DPA statement that the DAD6001 works with voltages down to 10 V per each line. In this case the current fed to the 4060 has to be very low, in fact if the 4060 is fed with a current of 0.15 mA, then the relative calculated voltage drop in the resistors is 7.26 V, which gives 2.74 V fed to the 4060. If this is true, then the 4060 can be fed with a very wide range of voltages and currents.

Considerations 2
It is clear that whatever calculation we do is not complete if we don't model the DPA4060 itself. As per my initial post, I made a few measurements of the 4060 in direct connection with the MT 24/96 (a bit risky, I would say now, but at the time I was just starting) as follows:
•   Test 1:   4060 connected to the MT 1/8" mic input (with plug-in power):
                               o   V =   2.52      Volts
                               o   I   =   0.9      mA      =>        Therefore Rmic= 2.800 kΩ
•   Test 2:   4060 connected to the MT 1/4" mic input (T+S) with phantom power on:
                               o   V=     2.592 Volts
                               o   I=       4.02  mA      =>   Therefore Rmic=  0.645 kΩ

In Test 2 it is likely that some form of internal protection intervened to limit the voltage, so I think that the current increase is not representative. Therefore, based on these tests, it could be deducted that the real operating voltage of the 4060 is around 2.5 Volts, with a current of 0.9 mA and a 4060 internal “resistance” around 2.800 kΩ.

Case 5
Repeating the calculations for the 4060 fed with a tension of 2.52 V and a current of 0.90 mA and considering the MT 24/96 (30V vs 48) feeding the same 0.90 mA to the load, i.e. 0.45 mA per each line of the balanced power supply (Tip and Ring), then the resistors total value (Rint+R2B+R1//R3) is 30.533 kΩ, which means that  each resistor R1, R2B and R3 adapted to the MT 24/96 phantom power has the value of 18.089 kΩ (instead of 30 kΩ  of the DAD6001).

Conclusions
Based on the above considerations, it seems to me the most sensible option to be with the DPA4060 fed by
a tension of 2.52 volts. In this case the current should be 0.90 mA.
Therefore the adaptor based on the DPA DAD6001 scheme modified for the 30 V phantom power provided by the MT 24/96 shall have:
o   All resistors:   18.089 kΩ (rouded up to 18.2 kΩ standard value)
o   Zener:      8.2 Volts (as a protection of the 4060)
o   All caps:   10 µF electrolytic as per DPA design (I will try to find a Polypropylene cap for the condenser C2, which gives the signal to the mic input, otherwise I will add a 100 nF bypass cap in parallel to the electrolytic one).
I’ll now build and test it. Hope it works, so I can “stealth” more easily my rig.
As a byproduct of this work, I think the MT implementation of phantom power is correct, maybe is not good for all mics, but for those requiring low current (and the DPA should be one of them) it should be OK.

Hope this is of interest and look forward to comments, contributions, etc.
Giovanni

PS: I can't insert the schematic of the DAD6001. Anyway it is available on the DPA website

[gorlando]

I built the adapter. Then, since I finally received the breakout cable from CoreSound, I tested it with my DPA4060 connected to the adapter, connected to the MT24/96 with phantom power on. Result:.... it is very noisy compared to the chain DPA4060 - CoreSound MIC2496 - MT24/96.
So this way is a dead end, unfortunately.

Then I built a simple battery box, sized to supply the same 2.5 volts / 1 mA, this resulting in 6.2 KOhm resistors, with 3.3 microfarad capacitors. I tested it with the DPA4060 and the 1/4" TRS MT24/96 inputs (without phantom power). Result:....very low noise!! and very good sound!! Bingo!! When I want a smaller rig I can leave home the MIC2496.

Then I thought that, since the 1/4" TRS inputs of the MT are designed to work with phantom power, there should be already the DC blocking condensers built in the MT itself. Therefore there should be no need of other condenser in the battery box. This not much for dimensional reasons but to avoid another condenser on the signal path from the microphone to the preamp.
To test this I simply by-passed the condensers of the battery box and...it works!
So the battery box (specific for the MT24/96) can consists only of the 9V battery and two 6.2 Khms resistors!

Now I am considering a final optimization: to take the power supply from the 1/8" input of the MT directly to the DPA4060. I tested this and measured 2.5 Volts with 0.9 mA, which is quite spot on what I think the DPA need. Since I still want to use for the signal the 1/4" TRS inputs of the MT (for sensitivity reasons) I will build a dongle connected to the 1/8" socket to provide power to the mics and to the 1/4" TRS input for the signal. If this works I will avoid the battery box completely and the rig will be very simple: DPA4060 - MT24/96. I'll do this in the next few days and report back the results.

I'll also put some pictures to show what I tried to descrive above.

Giovanni   

[it-goes-to-eleven]

That's some great analysis and effort, Giovanni.   I went the dongle route:

http://taperssection.com/index.php/topic,51031.0.html

[gorlando]

Freelunch, so you did it in 2005! I could have saved some effort and time had I known.
This forum confirms being a great site.
How does it perform? If the noise is low and the voltage & current sufficient, this is the ultimate stealth rig, consisting only of the mics and the MT, with eventually a battery extender for longer recordings.

Anyway, I post here some pictures of the phantom powered adaptor (unfortunately very noisy) and of the battery box (very good performance) I did.
Giovanni

[it-goes-to-eleven]

Very nice work. Maybe you'd be interested in the universal actives project, aka the AKG actives project?  Some schematics have been proposed and things are moving forward slowly..

http://taperssection.com/index.php/topic,59577.0.html

I have only run the dongle once. With dpa 4061's, the performance was fine for a Tool show.  I did not notice any distortion on the heavy bass.

Soon after receiving my microtrack I recorded a jazz show using the 4061's and a sound-pros 9v battery box.  I found the background noise of the MT to be somewhat high for that quiet source.

[gorlando]

I'm not sure I understand this "Actives" project. Is it to build a power supply for the AKG mic? Is this also including a pre-amp? I am not familiar with this terminology, my knowledge of electronics is because one of my hobbies is (was... I have little time now) to build HiFi equipment (pre, final amplifiers and the like). I have even built (25 years ago) a portable DBX which I used to tape concerts with a Sony TCD5M.
Anyway, coming back to the "actives", I'd like to understand what is the issue, so I can say if I can be of help.
Giovanni

[it-goes-to-eleven]

The intent is to drive the capsules without any bodies. Without the need for bodies, caps like the AKG and Gefell can be run in a hat and at fairly low cost.  The battery box may may or may not include a preamp depending on application.

Generally, the "actives" consist of a collette that mechanically connects to the capsule and a small powered FET circuit at the collette to drive the signal down the cable. At the opposite end of the cable is a 60volt source to polarize the capsule and another FET.  I know just enough to be dangerous but not enough to really innovate.  Fortunately it is pretty simple and even I can understand the circuit.

Eventually the collette and battery box should be fairly low cost.

[gorlando]

I don't have these AKG caps, and also I have no practice in mechanical works.
If you can address me to where I could find some schematics of the electronic part needed (the FET driver, the power supply), I could do something for this part, eventually also a prototype. However, not having the caps it would be difficult for me to test it.
In this scenario, I think I could add very little value to the efforts of such an expert group which is already working on the issue. In any case, thank you for the invitation.
Giovanni

[gorlando]

I completed the connector (see photos). The power to the mics is provided by the the 1/8" input of the MT, while the mic signal (too high for the 1/8" input of the MT) goes to the 1/4" inputs of the MT, which ara capable of handling the higher signal provided by the DPA4060.

I am also posting photos of my complete "stealth" rig. Now it is very small and at the same time very reliable (the connections are a custom made jack dongle, a positively locking mini-XLR and two microDot screwed connectors).
 
I have preliminarly tested this rig and it behaves like the battery box (see above in this topic), with very low noise and apparently no dynamic problems.

Now I will make some more realistic comparative tests among the three options I have (MIC2496, battery box and this MT direct feed) to see which is the best sonically (it should be the one with the MIC2496, but I want to check it) with what I normally tape (acoustic classical music concerts).

Hope the above can be of help to others
Giovanni
 

[Church-Audio]

I completed the connector (see photos). The power to the mics is provided by the the 1/8" input of the MT, while the mic signal (too high for the 1/8" input of the MT) goes to the 1/4" inputs of the MT, which ara capable of handling the higher signal provided by the DPA4060.

I am also posting photos of my complete "stealth" rig. Now it is very small and at the same time very reliable (the connections are a custom made jack dongle, a positively locking mini-XLR and two microDot screwed connectors).
 
I have preliminarly tested this rig and it behaves like the battery box (see above in this topic), with very low noise and apparently no dynamic problems.

Now I will make some more realistic comparative tests among the three options I have (MIC2496, battery box and this MT direct feed) to see which is the best sonically (it should be the one with the MIC2496, but I want to check it) with what I normally tape (acoustic classical music concerts).

Hope the above can be of help to others
Giovanni
 

Very nice work! I would consider using the 10k and a 10uf cap instead of the 2.2uf and then using a 250nF cap for a high pass filter that can be switched via a dpdt switch.
I would also lose the electrolytic caps.
Chris Church

[gorlando]

Chris,
The box with a lot of components (including electrolytics) is an adapter derived from the DPA DAD6001 (XLR Phantom 48V) to be used with the 30V phantom supplied by the MT. It works but is noisy, so I abandoned it.
However this exercise was useful because I understood from it that DPA feeds their 4060 with about 2.5V and 1 mA (with a mic internal "resistance" of about 2.8 kOhms).

I then built a battery box to provide the same voltage and current. Initially I used also capacitors (3.3 µF poly bypassed with 250 nF poly as well) and 6.2 k resistors, then, considering that the 1/4" MT input were designed to handle phantom power (i.e. there are already capacitors to block DC inside) I removed the capacitors, leaving only the resistors. It works, very low noise and good dynamics.

Finally I built an even simpler adapter which takes the power for the 4060 from the 1/8" input of the MT (this power is always on, even if other inputs are selected), so without any battery box.

Therefore I am using no additional capacitors (let alone electrolytics) on the signal line.

I don't know if I will need a high pass filter, I am taping mainly acoustic music taken in places with a very good acoustic (Concert / Opera Halls, etc.), so normally I don't have boomy low frequencies, but I will keep your suggestion in mind in case.
By the way, the filter parameters depend on the impedance of the MT input. Do you know the value of this impedance? I am guessing it should be at least 6 KOhm or more, but knowing the exact value (M-Audio does not publish it) would be of help.
Giovanni

[Church-Audio]

Chris,
The box with a lot of components (including electrolytics) is an adapter derived from the DPA DAD6001 (XLR Phantom 48V) to be used with the 30V phantom supplied by the MT. It works but is noisy, so I abandoned it.
However this exercise was useful because I understood from it that DPA feeds their 4060 with about 2.5V and 1 mA (with a mic internal "resistance" of about 2.8 kOhms).

I then built a battery box to provide the same voltage and current. Initially I used also capacitors (3.3 µF poly bypassed with 250 nF poly as well) and 6.2 k resistors, then, considering that the 1/4" MT input were designed to handle phantom power (i.e. there are already capacitors to block DC inside) I removed the capacitors, leaving only the resistors. It works, very low noise and good dynamics.

Finally I built an even simpler adapter which takes the power for the 4060 from the 1/8" input of the MT (this power is always on, even if other inputs are selected), so without any battery box.

Therefore I am using no additional capacitors (let alone electrolytics) on the signal line.

I don't know if I will need a high pass filter, I am taping mainly acoustic music taken in places with a very good acoustic (Concert / Opera Halls, etc.), so normally I don't have boomy low frequencies, but I will keep your suggestion in mind in case.
By the way, the filter parameters depend on the impedance of the MT input. Do you know the value of this impedance? I am guessing it should be at least 6 KOhm or more, but knowing the exact value (M-Audio does not publish it) would be of help.
Giovanni

I think its safe to say if its a line input its about 10k if its a mic input it could be anything from 150Ohm to 1k