Ribbon Mic Questions

[Ian VK3KRI]

Several Years ago I picked up a 'LustraPhone'  ribbon microphone. I believe it was manafactured in the UK as a consumer grade mic for tape recorders. 
Any way I decided I should do something with this, and hooked it up to a preamp. There is audio out, but not a lot of it and it seems a bit dead at high frequencies.  After some testing I discoverd that this actually has a 10K output impedance rathger than the lo Z output I was expecting ( Doh! The text 'High Impedance' on the nice enameled badge on the from should hage given that away) .  That explains the low output, now to the questions...

Is the high freq roll off likely to be due to
A the impedance mismatch ?
B the single  single transformer matching from the ribbon, say 0.2 ohms  to roughly 10K. A Z ratio about 50,000:1.  !
or
C The actual ribbon element itself?

I'd really like to get this mic to work as a normal low Z mic, but I suspect replacing the transformer may be a bit on the expensive side.  It may be easier to just build a tube preamp specifically for it.
                                       
                                                                  Ian VK3KRI

[Mike/W8BAC]

Hi Ian,

Have a close look at the ribbon first if you can. Usually the ribbon is protected by a cage of some sort. The powerful magnets on either side of the ribbon can grab that steel screw driver your using (or screws and cage your removing) will ruin your day if they are drawn in so be very careful.

Look for any debris stuck to the magnets that crosses over to the ribbon. Use a magnifier or visor. It is a common problem and the debris will dampen the ribbon and cause the type of problems your having. If the debris isn't serious you can use a toothpick to help clear it.

The ribbon is so delicate just thinking hard about it can ruin it. If it is jammed up it might be best to have a new ribbon installed by a pro. Good Luck.

Mike

[Rick K5IAR]

Ian,

The impedance mismatch is likely causing such low output you're not getting a true reading of the frequency response.  Have you considered an impedance matching transformer such as those sold by Heil or even Guitar Center?  I have used both and they work beautifully.  If you can get the mic operational you'll be very pleased with the excellent sound quality and frequency response.

Rick/K5IAR

[WBear2GCR]

Ian,

You did not say how you tested the mic, what you plugged it into??

"Preamp" does not tell us the input Z of the "preamp"!
If it is a low Z mic pre - 600ohm nominal variety - you are definitely swamping the thing. You need to try a high-Z input like the mic input of a consumer unit, tape recorder for example...

I would not try to reconvert the high Z transformer output using a transformer, not without a buffer. If ur intending to use it with a low Z input preamp, then consider the same FET follower trick that is frequently used with the D-104 mic - it is good enough for this sort of thing.

The other thing that can happen is that the magnets can lose their strength, especially Alnico that has been subjected to heat. The good part is that you can likely replace the alnico, especially if it is just a bar magnet or horseshoe type with some soft iron pole pieces and a hunk of Neodymium and get much more output due to increased flux.

How about a jpeg of this mic - especially if you take off the covers??

Fwiw, I strongly doubt any magnetic debris in the gap being the cause...

Oh, some mics of this sort have excessive padding and felt between the outside world and the ribbon! This was to protect the mic and of course reduce the HF response. A look under the screens will tell you that - or you might find that the back side is actually the front?? 

           _-_-bear

[Bacon, WA3WDR]

Low-Z microphones are supposed to go to open circuit preamps that give best noise performance with low-Z source impedances (amplifiers having more current noise than voltage noise).  But many designs put low-value resistors across them.

If you put a low resistance across a high-Z transformer coupled mike, the leakage inductance of the high-Z winding can work with the resistance to roll off the highs.

Bidirectional ribbon mikes (the ribbon between the magnet poles and open on both front and rear) tend to be bassy, especally close in, because they have serious proximity effect.  Properly connected, the top end is there, but the bottom is way out of proportion.  The mike may have a bass-cut switch for this reason.

And the ribbon often gets 'popped' and blown to one side, where it rubs against the magnet.  This can be cleared by very careful adjustment of the ribbon mount.  As was pointed out, the ribbon is nestled in the middle of a big magnet, and magnetic tools can get yanked away from you and tear things up.  If you try to repair it, I recommend strong reading glasses or a strong magnifier, so you can really see what's going on in there.

I've repaired ribbon mikes with ordinary hand tools.  I even made a new ribbon out of the foil from a cigarette pack once, or maybe it was from a chewing gum wrapper, I forget exactly.  That sort of stuff.  These things can be fixed.

[AB2EZ]

Ian
et. al.

If the ribbon is moving, the voltage across the ribbon is proportional to the velocity of the ribbon as it cuts through the magnetic lines of force between the permanent magnet poles.

If a sinusoidal pressure wave is applied across the flat face of the ribbon, the ribbon will respond by vibrating (and therefore producing a voltage) with an amplitude that is proportional to the amplitude of the pressure wave and inversely proportional to the frequency of the pressure wave. This is just a consequence of F=mA, where F is the force produced by the pressure wave, and m is the mass of the ribbon. The acceleration is proportional to the force... but the velocity (and thus the voltage produced) is the integral of the acceleration... yielding the 1/f dependency.

RCA, and later many others, compensated for this by placing an acoustic cavity around the ribbon... to try to make the pressure difference between the front of the ribbon and the back of the ribbon increase in proportion to the frequency of the incoming acoustic pressure wave. This was caused to happen by the intentional path delay, experienced by the incoming pressure wave, between the front and the back of the ribbon, within the acoustic cavity. For a given path delay, higher frequency pressure waves will experience more phase shift, and thus produce a larger pressure difference between the front and the back of the ribbon.

If your ribbon microphone does not have an acoustic cavity that creates the above effect, then the output will, indeed, drop off with frequency... thus requiring appropriate equalization. The good news is that the path delay effect actually rolls off low frequencies (rather than rolling up high frequencies)... so if you don't have the path delay, the lows will be boosted... but the highs will still be there.

The output of a typical ribbon microphone is very low compared to other commonly used microphones... and thus one would typically employ around 60-70 dB of gain in the preamplifier (v. about 40-50 dB of gain for a dynamic microphone).

I have a restored RCA 77DX ribbon microphone, and I use an extra 20 dB low noise "line amplifier" between that microphone and the preamplifier in my mixer. The low noise line amplifier has about 100k of input impedance and a low output impedance, so it provides not only 20dB of voltage gain.... but also a large current gain.

Stu

[WBear2GCR]

Replacement ribbon foil can be had by unwinding a film/foil capacitor.

Thin and fragile, but you get a whole bunch of it! 

            _-_-bear

[Ian VK3KRI]

I would not try to reconvert the high Z transformer output using a transformer, not without a buffer. If ur intending to use it with a low Z input preamp, then consider the same FET follower trick that is frequently used with the D-104 mic - it is good enough for this sort of thing.


How about a jpeg of this mic - especially if you take off the covers??

Fwiw, I strongly doubt any magnetic debris in the gap being the cause...

Here's the mic in question. There is a coarse weave sik around the ribbon, but you can see the corrugations in the ribbon in the 'dismantled' pic. The ribbon appears to be intact, not sagging or touching the sides.
I also discovered that wat I initially thought were insulating bushers were in fact 50 year old rubber grommets for shock absorbtion which are now solid! Also there's a neat ribbon tensioner thats not visibile in the pics - I don't think I'll be touching that.. ever.

I built a Hi Z FET buffer, and this seems to have resolved the freq response somewhat, unfortunately the buffer is rather noisy 
Whats the normal working distance for a fig-8 ribbon mic?  There is certainly a HUGE proximity effect when I get up close  , but when I back off, room echos become much more pronounced because of the resonse  off the back

Thanks for the feedback so far ..              Ian VK3KRI

[WD8BIL]

Hi Ian...... go to ur local guitar shop and pickup a "direct connect" box. These are used for older guitars with HI Z pickups. Usually TRS input and XLR output. Then ur Low Z preamp will work just fine.

[flintstone mop]

Hi Ian
Looks like an interesting mic. Ribbon mics are a little hard to get along with. Especially if it is old. There's good advice here to get it up and running. If you have some dollars the Marshall and other like it are great substitutes for the original ribbons and the high dollar condenser mics. In the USA we are paying around $99 for these fine mics. Getting into the broadcast types such as RE27's is a waste of money.
Good luck
Fred

[AB2EZ]

Ian

There does not appear to be any acoustic cavity type mechanism (as in the RCA ribbon mics, and many modern ribbon mics) to compensate for the normal 1/f response of a velocity (ribbon) microphone.

To get rid of the "proximity effect", which results from the basic physics of how this microphone works , I suggest that you add a 6dB/octave rolloff, starting at some adjustable low frequency... e.g., a simple RC filter.

Your ribbon microphone doesn't look a lot different from my homemade ribbon microphone:  http://mysite.verizon.net/sdp2/id16.html

The normal distance for talking into a ribbon microphone is fairly close... provided you never blow into it to test whether it is working.

Good luck!
Stu

[WBear2GCR]

Ian,

It looks like there are horseshoe type alnico magnets on the back? Two of them, one top and bottom??

Depending on the construction, IF the pole pieces have a means other than the magnets to hold them together, you could consider removing them, and replacing them with a hunk of neodymium magnet, or two neodymium magnets PER horseshoe magnet, with a machined bit of mild steel/iron completing the circuit. This would bump up the output considerably, as it would increase the flux by a considerable amount.

Of course, it is a classic peice, so maybe you just want to leave it alone...

As far as the FET "buffer"... if you use it for gain, maybe not as good!
You want to use it as a follower, and then use another xfrmr to bring up the voltage level... that should be close to dead quiet.

Paralleled FETs (or transistors) are quieter still.

But a good FET even used for gain should be very quiet.
I've used a 2SJ74 with a small triode in cascode as the front end for a MC cartridge RIAA preamp, and it is dead quiet. The total gain in the unit sans RIAA compensation is on the order of 90dB, before you go through the system preamp and then into the amplfier to the speakers... no sign of noise.

            _-_-WBear2GCR

[Ian VK3KRI]

Ian,

It looks like there are horseshoe type alnico magnets on the back? Two of them, one top and bottom??

Depending on the construction, IF the pole pieces have a means other than the magnets to hold them together, you could consider removing them, and replacing them with a hunk of neodymium magnet, or two neodymium magnets PER horseshoe magnet, with a machined bit of mild steel/iron completing the circuit. This would bump up the output considerably, as it would increase the flux by a considerable amount.

Of course, it is a classic peice, so maybe you just want to leave it alone...

As far as the FET "buffer"... if you use it for gain, maybe not as good!
You want to use it as a follower, and then use another xfrmr to bring up the voltage level... that should be close to dead quiet.

Paralleled FETs (or transistors) are quieter still.

But a good FET even used for gain should be very quiet.
I've used a 2SJ74 with a small triode in cascode as the front end for a MC cartridge RIAA preamp, and it is dead quiet. The total gain in the unit sans RIAA compensation is on the order of 90dB, before you go through the system preamp and then into the amplfier to the speakers... no sign of noise.

            _-_-WBear2GCR

Hmm.. A 'good' FET. I guess the used MPF102 I pulled out of the junk box may not qualify?  Actually the noise was due to using the FET for gain . I tried to use it as a phase splitter with equal resistance on drain and source, in turn driving a pair of transistors for balanced out.  Bypassing the drain resistor reduces the noise spectatuarly , but results in assmetric drive on the balanced line.  The far ends transformer coupled , so it shouldn't matter.

The magnets do appear to be removable, but I'll keep it in original condition.

                                                                        Ian VK3KRI

[WBear2GCR]

Ian,

I'm actually surprised that you got noise using the FET...

But, I'd not try to use it as a "phase splitter", and just use it as a current gain follower - like the schematics shown for a D-104 mod. Then follow that with a transformer and you've got ur phase split for balanced.

Probably the primary of the xfmr can be placed in the "source" line directly, no resistor needed.

You can connect the "drain" directly to the power supply line... the PS needs to be clean and noise free, so a battery might be best at least to start with.

FETs are self limiting as far as current flow with "0" bias on the gate... in fact they are used for current sources because of that aspect of their operation.

            _-_-bear