ELA M251 Inspired Microphone - Build Thread

[chunger]

Sorry for the recently broken photo links. Some forum migration may have broken them, and now, the forum is not allowing me to re-insert more than 30 photos per post. I am uploading documentation to my blog and will link there.

 

[chunger]

Build guide has been uploaded here with properly linked full photo set. Sorry I was not able to fix this thread in place.

ELA M251 Build Guide

 

[lightningbefore]

Thank you @chunger, was just going to ask!

 

[SubSpec]

Sooo, I've just dropped a CT12 into my JRR NOS 251, which is essentially this build with a 6021 (on the Matador PCB). It's always sounded decent but had a noise floor high enough that I don't reach for it often. It could just be a noisy tube (NOS GE 5-star), but rather than fool with swapping for other 6201's I'm leaning towards reverting it back to the 6072a configuration.

I need to check the components used in the JRR mic, but as I understand it the typical arrangement for switching from 6072a to 6201 would be dropping R17 to 47k. So assuming that's how it's configured, anything else I should take into account before swapping in a 6072a?

 

[Accelerator]

@SubSpec​

Changing the Anode Resistor (R17) 100KΩ to a 47KΩ anode resistor for an ECC81 - 12AT7 - 6201 tube works pretty well in a C-12 microphone which is a fixed-adjustable biasing microphone (so that with the 47KΩ anode resistor you can properly biasing it at -1.02V), but it doesn’t work as good for the Cathode Biasing ELA-M 251E microphone…

But let’s see the 47K case with the ECC81 - 12AT7 - 6201 tube:

120 DC Volts Vb / 47.000 Ohms = 0.0025531914…A aka 2.55mA so we have the 2.55mA as a marking point and we can draw our load line…

As we have the 120 DC Volts as Vb and we like to take 60 DC Volts as Quiescent Operating Point for Center biasing, we will see in the load line that at the 60 DC Volts the Grid Bias Voltage (V) is -1.02 Volts and the Iq(mA) is 1.28mA…

-1.02 Volts / 1.28mA = 796,875 Ohms, aka the nearest value is 800 Ohms for the Cathode Resistor (R1)…

The 800 Ohms is very low value for Cathode Resistor (R1) to performs well with the 22μF Cathode Bypass Capacitor (C1) a properly LC/HP Filter for a microphone, so in this case you have to increase the value of the Cathode Bypass Capacitor (C1) to 47μF or higher…

If you like to use the ECC81 - 12AT7 - 6201 tube in an ELA-M 251E microphone it is more preferable to keep the 100KΩ Anode Resistor (R17).

So let’s see this case, the 100K case with the ECC81 - 12AT7 - 6201 tube:

120 DC Volts Vb / 100.000 Ohms = 0.00120A aka 1.20mA so we have the 1.20mA as a marking point and we can draw our load line…

As we have the 120 DC Volts as Vb and we like to take 60 DC Volts as Quiescent Operating Point for Center biasing, we will see in the load line that at the 60 DC Volts the Grid Bias Voltage (V) is -1.58 Volts and the Iq(mA) is 0.60mA…

-1.58 Volts / 0.60mA = 2.630,333333 Ohms aka the nearest value is 2.630 Ohms for the Cathode Resistor (R1)…

So you can use the Vishay Dale CMF55 2.61k Ohm, 1% as Cathode Resistor (R1) and with this value it will work pretty well in combination with the 22μF Cathode Bypass Capacitor (C1) to performs a properly LC/HP Filter for a microphone, so in this case you don’t have to increase the value of the Cathode Bypass Capacitor (C1) to 47μF or higher…

Keep in your mind that back in time the AKG AG used the 100KΩ Anode Resistor – 2.7KΩ Cathode Resistor for the legendary C-24 microphone, but today we can have more precision in today’s value choices…

 

[SubSpec]

Thanks @Accelerator!

Going to simplify and just revert the circuit back to the standard 6072a setup. But, thought I'd note how the JRR mic is configured, as it caught me off guard. R17 is 180k and C12 is a 3.3uf and 3.5uf in series :/

So, I'll dive into the PSU before going any further. I'm guessing this was an attempt to use an off the shelf power supply, but seeing as I have little idea what I'm doing in general that's nothing more than a guess. Should be easy to find the answer to though tomorrow...

Edited to add:
Now it's tomorrow, and there is definitely problems coming from the PSU with 1.1v and 0.5v respectively coming off of the audio pins straight off the PSU. For context, I had swapped the resistor from the 180k to the prescribed 47k (for now, again the plan is to swap back to 100k and a 6072a once I get one). Instantly the noise got far more prominent and changed from a fairly constant low frequency noise to something more modulated and high frequency. So I plan to dive into the PSU (haven't so much as opened it yet) to see how it's configured and will likely end up getting a Matador PSU board as I am certain this is a stock PSU.

 

[pasarski]

@SubSpec​

Changing the Anode Resistor (R17) 100KΩ to a 47KΩ anode resistor for an ECC81 - 12AT7 - 6201 tube works pretty well in a C-12 microphone which is a fixed-adjustable biasing microphone (so that with the 47KΩ anode resistor you can properly biasing it at -1.02V), but it doesn’t work as good for the Cathode Biasing ELA-M 251E microphone…

But let’s see the 47K case with the ECC81 - 12AT7 - 6201 tube:

120 DC Volts Vb / 47.000 Ohms = 0.0025531914…A aka 2.55mA so we have the 2.55mA as a marking point and we can draw our load line…

As we have the 120 DC Volts as Vb and we like to take 60 DC Volts as Quiescent Operating Point for Center biasing, we will see in the load line that at the 60 DC Volts the Grid Bias Voltage (V) is -1.02 Volts and the Iq(mA) is 1.28mA…

-1.02 Volts / 1.28mA = 796,875 Ohms, aka the nearest value is 800 Ohms for the Cathode Resistor (R1)…

The 800 Ohms is very low value for Cathode Resistor (R1) to performs well with the 22μF Cathode Bypass Capacitor (C1) a properly LC/HP Filter for a microphone, so in this case you have to increase the value of the Cathode Bypass Capacitor (C1) to 47μF or higher…

If you like to use the ECC81 - 12AT7 - 6201 tube in an ELA-M 251E microphone it is more preferable to keep the 100KΩ Anode Resistor (R17).

So let’s see this case, the 100K case with the ECC81 - 12AT7 - 6201 tube:

120 DC Volts Vb / 100.000 Ohms = 0.00120A aka 1.20mA so we have the 1.20mA as a marking point and we can draw our load line…

As we have the 120 DC Volts as Vb and we like to take 60 DC Volts as Quiescent Operating Point for Center biasing, we will see in the load line that at the 60 DC Volts the Grid Bias Voltage (V) is -1.58 Volts and the Iq(mA) is 0.60mA…

-1.58 Volts / 0.60mA = 2.630,333333 Ohms aka the nearest value is 2.630 Ohms for the Cathode Resistor (R1)…

So you can use the Vishay Dale CMF55 2.61k Ohm, 1% as Cathode Resistor (R1) and with this value it will work pretty well in combination with the 22μF Cathode Bypass Capacitor (C1) to performs a properly LC/HP Filter for a microphone, so in this case you don’t have to increase the value of the Cathode Bypass Capacitor (C1) to 47μF or higher…

Keep in your mind that back in time the AKG AG used the 100KΩ Anode Resistor – 2.7KΩ Cathode Resistor for the legendary C-24 microphone, but today we can have more precision in today’s value choices…

I find it really difficult to draw load lines on a typical triode curves with very good precision at typical microphone operating region. Would you oppose if I interpreted the calculations as a starting point and adjusted the final values by ear? There must be enough variation between tubes also to have some margin regarding the actual resistor values, say 2.7k Rc for an other tube and a 2.8k for an other specimen? Also, is there not a also margin to adjust the values for personal preference about how linear the mic should sound?

 

[Accelerator]

You welcome @SubSpec

With 120 DC Volts as Vb and 47KΩ as Anode Resistor you can also have the alternative option to use the ECC82 - ECC802 - E82CC - 12AU7 – 5814A – 6189W tube…

So let’s see the 47K case with the ECC82 - ECC802 - E82CC - 12AU7 – 5814A – 6189W tube:

120 DC Volts Vb / 47.000 Ohms = 0.0025531914…A aka 2.55mA, so we have the 2.55mA as a marking point and we can draw our load line…

As we have the 120 DC Volts as Vb and we like to take 60 DC Volts as Quiescent Operating Point for Center biasing, we will see in the load line that at the 60 DC Volts the Grid Bias Voltage (V) is -2.55 Volts and the Iq(mA) is 1.28mA…

-2.55 Volts / 1.28mA = 1992,1875 Ohms aka the nearest value is 2K Ohms for the Cathode Resistor (R1) and you can use the Vishay PR01-2k 1W resistor…

Yes, I already know that there is a huge “drop down” in the μFactor(s) from the μFactor = 40 that the 6072A – 12AY7 tube have to the μFactor = 18 that the ECC82 - ECC802 - E82CC - 12AU7 – 5814A – 6189W tube have, but let me remind you that the legendary AC701K tube that the legendary AKG-Telefunken ELA-M 250 (non E) and the Neumann M-49 (a,b,c) microphones used, also had μFactor = 18 and the lower Ra (Rp for the Americans…) that the ECC82 - ECC802 - E82CC - 12AU7 – 5814A – 6189W tube has makes it a better “matching” with the Output Transformer, aka in this case you will have a more “Neumann-ique” approach…

But if you like to use the “classic” approach of the 6072A – 12AY7 tube in an ELA-M 251E microphone, it is more preferable to keep the 100KΩ Anode Resistor (R17).

So let’s see this case, the 100K case with the 6072A – 12AY7 tube:

120 DC Volts Vb / 100.000 Ohms = 0.00120A aka 1.20mA, so we have the 1.20mA as a marking point and we can draw our load line…

As we have the 120 DC Volts as Vb and we like to take 60 DC Volts as Quiescent Operating Point for Center biasing, we will see in the load line that at the 60 DC Volts the Grid Bias Voltage (V) is -1.23 Volts and the Iq(mA) is 0.60mA…

-1.23 Volts / 0.60mA = 2.050 Ohms aka the nearest value is 2.05K Ohms for the Cathode Resistor (R1)…

So you can use the Vishay Dale CMF55 2.05k Ohm, 1% as Cathode Resistor (R1) and with this value it will work pretty well in combination with the 22μF Cathode Bypass Capacitor (C1) to performs a properly LC/HP Filter for a microphone, so in this case you don’t have to increase the value of the Cathode Bypass Capacitor (C1) to 47μF or higher…

 

[Accelerator]

@pasarski​

I find it really difficult to draw load lines on a typical triode curves with very good precision at typical microphone operating region. Would you oppose if I interpreted the calculations as a starting point and adjusted the final values by ear? There must be enough variation between tubes also to have some margin regarding the actual resistor values, say 2.7k Rc for an other tube and a 2.8k for an other specimen? Also, is there not a also margin to adjust the values for personal preference about how linear the mic should sound?

It’s your microphone Man…

You can biasing “colder” if you like so, you can also biasing it “hotter” if you like is so, and no-one is here to “judge” you for your likes and don’t forget that in all the history of the tube microphones several technical “mistakes” or uncorrected technical choices that all the legendary factories (AKG, Neumann, Schoeps) has done becomes “Happy Accidents”…

Even the Sony C-800G microphone has several technical mistakes, but guess what, it becomes a legendary microphone for the people that they like it…

 

[pasarski]

@pasarski​

It’s your microphone Man…

You can biasing “colder” if you like so, you can also biasing it “hotter” if you like is so, and no-one is here to “judge” you for your likes and don’t forget that in all the history of the tube microphones several technical “mistakes” or uncorrected technical choices that all the legendary factories (AKG, Neumann, Schoeps) has done becomes “Happy Accidents”…

Even the Sony C-800G microphone has several technical mistakes, but guess what, it becomes a legendary microphone for the people that they like it…

Thanks What about the variation of tube specimens? I don't have a clue how much they vary. Lot less than fets is all I know. Also, doesn't the heater voltage affect the behavior? If the tube is underheated the "correct" bias point changes a little bit?

 

[Accelerator]

Thanks What about the variation of tube specimens? I don't have a clue how much they vary. Lot less than fets is all I know.

Neither do I can I tell you or be able to know about all the variation of tube specimens, nor do I have any clue how much they vary from the one tube factory to the other tube factory, neither from the one brace to the other brace from the same factory in time, because as you know the Amplitrex AT1000 it is not cheap as cheeps ($8,052.32 U.S.D.) and I don’t have neither the money to investigate them in buying this instrument just for metering electronic tubes / valves, neither do I have a titanic amount of all the pre-amp. electronic tubes / valves of all the types and the brands that they have been used in all the microphones and electronic equipment(s)…

Having the Amplitrex AT1000 for metering electronic tubes / valves and a titanic amount of all the pre-amp. electronic tubes / valves of all the types and the brands that they have been used in all the microphones and electronic equipment(s) will be a “wet dream” for every tube/valve DIY-er, but as you know such as “wet dreams” are too expensive for every one, while the online loadline calculator(s) are free…

If the tube is underheated the "correct" bias point changes a little bit?

If the tube / valve is under-heated the "correct" bias point and the whole “characteristics” changes a lot and this is also one of the reasons about the “mythical” Neumann U-47 “sound”…

Read this article by the Croatian Moxtone Labs about how much the under-heated change the "correct" bias point and the characteristics of the VF14, EF14, EF13, EF12 tubes…

A JOURNEY INTO THE LEGEND: THE OUTPUT CHARACTERISTICS OF THE VF 14 AND ITS SUBSTITUTES

But as I had wrote above and as you guess the “mythical” Neumann U-47 “sound” is one of the several technical “mistakes” or uncorrected technical choices that all the legendary factories (AKG, Neumann, Schoeps) has done and becomes “Happy Accidents”…