Hi:
I have some 'relatively deaf' dynamic percussion mics that are probably 20-30 dB less sensitive than the other dynamic mics I have. I don't recall the sensitivity reference units, so forgive me...with the same unstated sensitivity reference, I have some that are ~ -75 dB(?) and -86 dB(?) vs. -56 dB(?) for the vocal mics.
They are all in the 150 - 600 ohm range dynamic mics.
Occasional home recording of amateur musicians through the low budget recording gear I'm kludging together is the target use...
Years ago I bought some EF86's and found an Eric Barbour schematic on the old Svetlana site that has been archived on the German site Tube Town...
http://www.tube-town.net/diy/sv-ef86/sv-ef86_01.jpg
I haven't done any math, but it looks like it potentially has enough gain, but I don't know if that remains true across the wide range of input impedance applications summarized in the schematic notes...two extremes being a dynamic mic vs a condenser (which I don't have but the possibility of a hi-z tube preamp for a piezo instrument pickup is intriguing...)
I don't know what kind of output coupling I'll be doing...they recommended minimum 500k ohms. I have more 5867's lying around than anything else...maybe a cathode follower of some sort after the EF86...I do have an overabundance of power transformers, so that's not an obstacle. I have a couple 50/200: 50k Newcomb TR-91 transformers but they violate the 500k recommended load for the EF86. They are going to be a PITA - 9 pin sockets not easy to find, and why waste them on a Behringer musicl instrument mixer anyway?
I'm wondering how this circuit looks regarding noise in this application I've proposed...any advice is appreciated...
Thank you
Murray
I have some 'relatively deaf' dynamic percussion mics that are probably 20-30 dB less sensitive than the other dynamic mics I have. I don't recall the sensitivity reference units, so forgive me...with the same unstated sensitivity reference, I have some that are ~ -75 dB(?) and -86 dB(?) vs. -56 dB(?) for the vocal mics.
They are all in the 150 - 600 ohm range dynamic mics.
Occasional home recording of amateur musicians through the low budget recording gear I'm kludging together is the target use...
Years ago I bought some EF86's and found an Eric Barbour schematic on the old Svetlana site that has been archived on the German site Tube Town...
http://www.tube-town.net/diy/sv-ef86/sv-ef86_01.jpg
I haven't done any math, but it looks like it potentially has enough gain, but I don't know if that remains true across the wide range of input impedance applications summarized in the schematic notes...two extremes being a dynamic mic vs a condenser (which I don't have but the possibility of a hi-z tube preamp for a piezo instrument pickup is intriguing...)
I don't know what kind of output coupling I'll be doing...they recommended minimum 500k ohms. I have more 5867's lying around than anything else...maybe a cathode follower of some sort after the EF86...I do have an overabundance of power transformers, so that's not an obstacle. I have a couple 50/200: 50k Newcomb TR-91 transformers but they violate the 500k recommended load for the EF86. They are going to be a PITA - 9 pin sockets not easy to find, and why waste them on a Behringer musicl instrument mixer anyway?
I'm wondering how this circuit looks regarding noise in this application I've proposed...any advice is appreciated...
Thank you
Murray
If Rin is set per the instruction, then I would think the gain and frequency response should be as specified.
As long as the load impedance is within reason, the EF86 should have no problem driving it, so you do not need to have exactly a 500k load as recommended. The Newcomb TR-91 is typically used as an input step-up transformer, so you can use it to get more gain if needed, it may also be ok wired backwards as an output transformer, but like you said, why waste it on a Behringer
BTW, the 20/500:50k specification merely establishes the turns ratio, it does not mean that one side is 20/500 Ohm and the other side is 50k. A transformer reflects the impedance connected to it from one side to the other side. Wiki here.
Thank you...for the reply and reading my meandering post...
Today I remembered the option of stacked jfet's in formats I'm not recalling the names of right now...maybe mu-follower, SRPP, cascode...and a JFET-bipolar combo.
I wonder if a mu-follower is a simpler option...thinking about them in the context of guitar pickup buffers or preamps, I had dismissed them, thinking 'what would I possibly do with such high gain when I want a buffer'?
Now I can use some...
Today I remembered the option of stacked jfet's in formats I'm not recalling the names of right now...maybe mu-follower, SRPP, cascode...and a JFET-bipolar combo.
I wonder if a mu-follower is a simpler option...thinking about them in the context of guitar pickup buffers or preamps, I had dismissed them, thinking 'what would I possibly do with such high gain when I want a buffer'?
Now I can use some...
When I was building with EF86s… yes, I know. But we would always have put an input transformer in front of a stage like that, particularly with a low sensitivity mic, and you have one available. 45dBs is not a lot of gain, and the transformer would give you a massive boost, with a minimum of extra noise (with the 620Ω across the input you're shorting out some of the theoretical input noise, but the tube's self generated goes on). You could put an input switch so as to have a high gain tube DI, a transformer input or an unbalanced loZ input, and there's no reason why its usefulness stop at those amateur musicians.
A simple cathode follower is probably the easiest output; you've got the voltages, it'll have the impedance and drive necessary. Yes, it's (cringe) unbalanced, but unless you've got a designed line level output transformer (using a mic transformer backwards is not only risking bass distortion, but throwing away a mass of that gain you were looking for), it's the most practical reply.
Put in some form of attenuator (Yay! I spend all that time fighting for gain, then talk about throwing it away, but tubes can produce enormous levels before distortion, Behringers can't always accept them) if it were me, I'd have a switched pad before the transformer, and, assuming you're biassing the cathode follower from the anode of the EF86, a series resistor here and a resistor/capacitor to ground. Possibly, if you can get a sufficiently complex multipole switch out of your junk box (tube designs used to use them a lot) remove the cathode bypass resistor from the pentode at the same time.
And now my diy soul starts adding high pass filters. powering for external tube microphones, a DC biassed input for Cducers (assuming they exist any more)… ignore me. It's senility setting in.
A simple cathode follower is probably the easiest output; you've got the voltages, it'll have the impedance and drive necessary. Yes, it's (cringe) unbalanced, but unless you've got a designed line level output transformer (using a mic transformer backwards is not only risking bass distortion, but throwing away a mass of that gain you were looking for), it's the most practical reply.
Put in some form of attenuator (Yay! I spend all that time fighting for gain, then talk about throwing it away, but tubes can produce enormous levels before distortion, Behringers can't always accept them) if it were me, I'd have a switched pad before the transformer, and, assuming you're biassing the cathode follower from the anode of the EF86, a series resistor here and a resistor/capacitor to ground. Possibly, if you can get a sufficiently complex multipole switch out of your junk box (tube designs used to use them a lot) remove the cathode bypass resistor from the pentode at the same time.
And now my diy soul starts adding high pass filters. powering for external tube microphones, a DC biassed input for Cducers (assuming they exist any more)… ignore me. It's senility setting in.
I have not dealt with enough mics or mixers to know... but doesn't the input sensitivity adjustment on even the cheapo Behringer offer enough gain/boost for even for the least sensitive dynamic mics? Why do you need an external mic-preamp, I thought those were meant for condenser mics with very low outputs...
Cool ideas - thanks, both.
The mu-follower idea revisited me as I was thinking about needing significant gain for the percussion mics for non-percussion use...if viable with the low-Z dynamic mics, it would be quite a bit simpler than an EF86...and certainly quicker to determine success or failure...
The other thought was about the EF86 Svetlana circuit with the high-Z input configuration for a piezo...but I would not need the gain that circuit offers...a hi-Z buffer like even a JFET would suffice, so just cross the piezo distraction off the list of questions!
I might try a mu-follower out of curiosity...with the low-sensitivity mic...
The mu-follower idea revisited me as I was thinking about needing significant gain for the percussion mics for non-percussion use...if viable with the low-Z dynamic mics, it would be quite a bit simpler than an EF86...and certainly quicker to determine success or failure...
The other thought was about the EF86 Svetlana circuit with the high-Z input configuration for a piezo...but I would not need the gain that circuit offers...a hi-Z buffer like even a JFET would suffice, so just cross the piezo distraction off the list of questions!
I might try a mu-follower out of curiosity...with the low-sensitivity mic...
If you are, say, distant miking a small ensemble with ribbon mics (something I've done many times, including with tube preamps) you need vast quantities of gain - sometimes in the order of 70-75 dBs. This means that electronic noise overtakes acoustic, and while I'm not the stickler for noise of someone from the digital generation, I still like the music louder than the hiss
.Condensor mics generally give more level than dynamic (using a power supply and modulating rather than generating), so remarkably few small consoles have the spare gain; mostly you find it in outboard preamplifiers. Generally, if you see "60dB gain", you discover this is not the onboard mic preamp, but mic input to stereo output with all faders at max. Not optimal for noise. Transformers are a great way of amplifying signal with minimum increase in noise, but are not over fashionable (despite frequently sounding very good indeed – at least, with decent transformers).
Behringer (oh, not only them, but they're on this thread already) like to make their specifications, particularly on noise figures, look the best possible, just about theoretical; so they don't want too much gain. Also, because they tend to trim headroom for the same reason, they don't want too much level coming out of their preamps, either.
They're still a darn sight better than the mic preamps built into computers.
I'm back to the EF86 because it looks so simple.
I looked at Aikido Cascode and Geek's Cascode pramp as well...I have a handful of EF86's and a lot of 5867's. I would have to change some things because of the operating point of 5687's. I looked at the EF86 Barbour mic preamp configurable option and it is SO simple it's hard to resist...just need power supply.
I also have alot of 6080's. I found John Broskie's Cascode Feedforward/Feedback Shunt Regulator using a cascode 6080, and it too is remarkable simple. Other than the largish filament current the 6AS7/6080 needs...it's simple and I have most of the parts already.
Some say there is no point in regulating a plate supply for a pentode (talking about the EF86), but regulator would be a way to get the desired voltage rather than seeking some diminishing benefit.
I just found a recycled chassis that has 4 9-pin sockets and space for 2 octal sockets and that looks inviting.
The question is - why would I bother to go a step further than the cascode 6AS7 to a pentode when regulating a supply for a pentode amp is looked down on anyway?
The Barbour Svetlana app note discusses the input resistance configuration as needed, including 620 ohms for a 1:1 input transformer. I have 1:1 transformers, so unless a pentode being driven by a low-Z input has some horribly inappropriate feature that eludes me, this may be my simplest solution to a better mic preamp front end.
If regulating a supply for a pentode is really unimportant, dispensing with the 6AS7 filament baggage, the EF86 plate current is tiny and the filament current is also rather low (0.2A @ 6.3V).
For modest dynamic mics and maybe a somewhat 'better' lavalier electret mic I have sitting around, I may just put this on my schedule...I have no plans for anything fancier like condenser or ribbon mics...at present.
Unless there is an opinion that I missed about the EF86 mic stage (odds of a non-microphonic EF86 in my basement?)...
Thanks.
I looked at Aikido Cascode and Geek's Cascode pramp as well...I have a handful of EF86's and a lot of 5867's. I would have to change some things because of the operating point of 5687's. I looked at the EF86 Barbour mic preamp configurable option and it is SO simple it's hard to resist...just need power supply.
I also have alot of 6080's. I found John Broskie's Cascode Feedforward/Feedback Shunt Regulator using a cascode 6080, and it too is remarkable simple. Other than the largish filament current the 6AS7/6080 needs...it's simple and I have most of the parts already.
Some say there is no point in regulating a plate supply for a pentode (talking about the EF86), but regulator would be a way to get the desired voltage rather than seeking some diminishing benefit.
I just found a recycled chassis that has 4 9-pin sockets and space for 2 octal sockets and that looks inviting.
The question is - why would I bother to go a step further than the cascode 6AS7 to a pentode when regulating a supply for a pentode amp is looked down on anyway?
The Barbour Svetlana app note discusses the input resistance configuration as needed, including 620 ohms for a 1:1 input transformer. I have 1:1 transformers, so unless a pentode being driven by a low-Z input has some horribly inappropriate feature that eludes me, this may be my simplest solution to a better mic preamp front end.
If regulating a supply for a pentode is really unimportant, dispensing with the 6AS7 filament baggage, the EF86 plate current is tiny and the filament current is also rather low (0.2A @ 6.3V).
For modest dynamic mics and maybe a somewhat 'better' lavalier electret mic I have sitting around, I may just put this on my schedule...I have no plans for anything fancier like condenser or ribbon mics...at present.
Unless there is an opinion that I missed about the EF86 mic stage (odds of a non-microphonic EF86 in my basement?)...
Thanks.
Well for your information, the 20/500:50k specification merely establishes the turns ratio, it does not mean that one side is 20/500 Ohm and the other side is 50k. A transformer reflects the impedance connected to it from one side to the other side. transformer helping article here
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.