A lucky day.
Before 2000 a pair of those mics sitting on a table inside their violet lined wooden box were found. It must had been a Vintage Wireless Society, or a HiFi fair in England, NEC Birmingham?. The 2 young persons on the stand gave them for a very reasonable price.
It was first time touching a ribbon mic.
They might had been mistreated bedore or after with ohm measuring abuse. It can tear the ribbon away very easily. Or the ribbons were already damaged or there were bad contacts. Sound output was so low. A company was found (the name might have been John or J. Laird or Leard and Sons? gold beaters, Endinburgh?). They kindly provided silver and aluminum foil of such a small thickness. It was an exercise how to handle it. It was decided to breath by a straw so that air does not make the material fly. First silver was tried. But then aluminum. It was randomly corrugated by hand. The ribbon was installed while listing to the sound of the mic. It was confirmed what was read in books that the reasonant frequency must be at the low end.
The ribbons were fastened on the brass supports which had previously been polished as you would silverware.
The mics were nicely operating.
But difficult mics, they were difficult to get an acceptable signal to noise ratio. 25 ohms was too low.
When singing close the bass was so huge and unnatural.
But there was always the belief that these mics are fabulous after reading in all those British and non British publications that ribbon mics are state of the art in transient response due to the low mass of the moving element.
The first setup that worked really well and it is possible the best was when Brian Sowter made a custom 1:20 input transformer to take advantage of the 25ohms output impedance.
The secondary was directly feeding the input of the Ferrograph series 6 electronic valve (electron tube) reel to reel recorder running at 7.5ips. The same ones used by BBC and there are photos of John Lennon using them too.
The mics were used as Blumlein pairs. They were used in churches for guitar and flute recitals. In halls for recording big classical orchestras. The sound was amazing. And the conductor Toufeksis who conducted Gershwin called to say congratulations and how the sound reminded of the great 50s, 60's American recordings.
But on closer singing the proximity effect was taking over.
And there was always some hiss which made things far from perfect on softer acoustic sources.
It was decided to try the Coles 4038. It was perhaps the first email ever sent. And it was Alan Parson's himself at APRS? replying that he used them on Dark Side of the Moon as overheads. When they came they were very nice but little was known of the proximity effect. And the Coles of course sounded much more bass heavy.
The conductor did not call again to say congratulations as a Steinway piano concert was recorded with the 2 4038 without proximity compensation and the sound was bass heavy or treble lacking.
Panayotis said he does prefer the Grampians as they sound more bright.
He was not believed.
Not long before the Grampians were used to record a rock band in the living room and the result was amazing.
But the 4038 should be better and every conceivable experiment was done on how to compensate for the proximity effect while making very low noise valve preamps for both Grampian and Coles mics in quest of a perfect sound with no hiss.
Laminations from the input transformer started to be removed to compensate for the proximity effect by reducing the bass response.
Electron tube preamps were built with E88CC long tailed pairs. The input transformer was a Sowter 1:7 designed for 200hm mics. It was mainly used with the 4038.
But Panagiotis gave a great schematic of a single K117 JFET and said why don't you connect the ribbon directly to the gate.
He must had been jocking!
But this was a great idea for a pre preamp with an input transformer. Inside of the mic you have the first step up transformer. Then the usually outside step up input transformer feeds the K117.
So a small box was made with a Jensen modified 1:10 transformer and a K117 JFET. This is the first Pleiades (transistor) JFET pre pre preamp so it is called T1. The Jensen was so much modified by reducing the primary turns that the inductance had become only 4.7mH and the step up ratio 1:60?. Many laminations were removed too to get the input inductance this low.
And it was connected to an almost broken Sennheiser MD421 black, which had a problem of sounding very thin the first few seconds of minute.
The sound was very bright. But of course low noise.
But one day the day the T1 was connected to the Grampian GR1/L and the sound started to be very interesting, amazing, and hiss going really low. And of course one could be singing almost next to the ribbon with no problems of pop because of electromagnetic damping of the low input impedance at low frequencies of the modified input transformer.
Many preamplifiers were designed and built at that time.
A 2 stage, 2 channel 6072 electron tube preamp with input, interstage, output transformers, long tailed pair. The schematic looking similar to an RCA BA6A but simpler. Everything operating with 4x 12 volt batteries so that phantom power voltage is ready too if needed. And low noise since the batteries have such a pure DC voltage.
Another 2 channel prepreamp with one UCC85 per channel was made again with no coupling capacitors. Long tailed pairs. Similar to the first stage of an RCA BA6A only simpler. It was nice as the high heater voltage would also supply the anode circuits. So again battery powered. No control room available just listening with headphones which were the Beyerdynamic DT770 which sounded bright.
After a break with the rock band recording while playing it was decided to change the tubes to E88CC by adding a series resistor to the heater circuit. The wooden floor was then found with a black spot from the heated up resistor touching it. It was observed at that time that flashing light would be heard as sound. There was no grid return circuit. Each grid was connected to one of the input transformer secondary wires. The flashing light from switching on fluorescent lamps was possibly photo emitting electrons? Years after the gain of this amp modified for ECC82 was found again to change with light, an agc compressor idea?
This amp sounds really nice and back then was connected to the AKG C12A in Blumlein pair feeding a Philips CDR recording the live rock band.
More experiments with ECC86 directly operating with 6 volts for both anode and heaters. But never too impressive.
A T2 K117 version was made with a military transformer of just 47mH primary inductance to compensate for the proximity effect. Lefteris advice was important, to use military transformers. And it was a Telefunken AM-75 GRC military amplifier which had an even less input inductance making a very bright and interesting sound.
It was decided to shunt the 47mH by another inductor to lower the input inductance and perhaps this is how the Pleiades filter were born.
The resultant inductance was made to 4.7mH.
Such little pre preamps were used to drive the Tascam porta one and two recorders with very pure sound and very low noise.
Somewhere around that time it was read while Google searching for battery electron tube operation as front end amplifiers for radio telescopes. On Jeff Duntemann's Junkbox website it was read that a so called space charge tube configuration exists that makes power amplifier operation of an electron tube with just 12 volts. By applying the 12V to the first grid of the space charge tetrode. Then the input signal to the other grid.
Having previous experience with one transistor circuits driving a loudspeaker, circuits by Jones, and Clive Sinclair referred to on other euroelectron posts. It was almost immediate that the Pleiades 1 electron tube headphone amplifier was conceived and made. The Pleiades Electra 2. 2 tubes for stereo driving the Sennheiser HD580 studio high Z headphones. No other components.
While listening to its great sound the sound kept improving as the 12V battery was discharging.
So by chance it was found that underheating makes a miracle.
It was also remembered on an old phone conversation a RIAA preamp experimenter in Athens mentioned that he uses underheating for low noise record player cartridge preamplifiers.
A V14 Neumann circuit of U47 had already been built by suggestion of Lefteris. Initially it was powered by 4 12V batteries. And the transformer was directly connected to the anode. It sound fantastic.
But then it was decided to built the exact circuit. So and anode resistor was added, tube voltage regulator and a normal power supply with rectifier tube so that it can be connected to the mains. Everything was built inside an nice 2U box bought from Canford in UK.
There was some very little hum and the Telefunken output transformer had to mounted a bit at an angle!
During that time it must had been observed too that the VF14 is operated by George Neumann at lower that the manufacturer specified heater voltage!
While teaching Hliana how to play the piano by ear brain based on what is called now Bach's Bass Rules PDF, one day it was decided to explain to her how electron tubes operate. So the wall of a small room at our back become the cathode to our imagination. The wall in front to us the anode.
Somewhere in between was the grid and it had to be biased negatively as this is what all the textbooks teach.
Hliana who had just read an article possibly by Montalk on how negative bias is used to control people was very strict about it.
"No, the grid must be positive to free the electrons."
At this time experimentation was also made with large grid resistors in order to make tubes behave as compressors by the rectification effect as is done on the Neuman U47 circuit in large input signals. And the high resistor insured a high release time until the grid becomes from more negative to less negative returning to its normal state.
6BC8 2 stage tubes were tried with few volts but perhaps not too great results although the sound was loud and fine.
Somewhere before that time electrometer tubes were also tried by the suggestion from Sawn Davies in UK .
So the 47mH military transformer was connected to an electrometer tube through a capacitor. The heater voltage was a 1.2V battery. The anode voltage was from a 9 volt battery through an anode resistor.
One day by mistake the Vb had been removed and the Vb terminal was connected to ground. So instead of a the battery a short circuit was connected. And the pre preamp operated, amplified sound!
It took a little time to realize that the 1.2V battery was playing both the part of heater and anode voltage.
Then when the grid return high Megohm resistor was connected to the other heater terminal an increase of anode current and gain was observed. The grid was less negatively biased this way. This amp is called V1 as it is not the first Pleiades electron tube with just a battery.
There are many more but before 2000 a small one electron tube ECC86 was connected in a very small aluminum die cast box to a military output transformer. It never worked white impressively but now with the updates it is proudly called V-1 and has an ECC82 instead. It is described on a dedicated euroelectron post.
Mr Giannis who is above 90 years old mentioned that he really likes changing an EF89 tube to the EF183 in radios.
It was decided to try various tubes including the EF37A used in the EMI RS61. The 6SK7 used in RCA BA6A, EF86 and EF183 on those Pleiades low voltage pre preamp expeeiments.
It was always a paternal suggestion that experiments should be done with batteries for safety.
Great idea!
Some experiments were done with the above mentioned tubes and high Megohm grid resistors to ground. By that time the V-1 was improved by connecting the cathode to ground and adding a high Megohm resistor from cathode to grid inspired by the low noise Neumann circuit. The tube now was ECC82 as it was found to operate better than the ECC86, E88CC etc.
Of course the cathode resistor was connecte to ground as it seems logical to have as much current from the tube without the negative bias created by the cathode resistor.
A few years ago a then abandoned pre preamp was made with an 7025, just one half. The anode voltage was with just 2 9V batteries in series. And one of them was feeding the heater circuit. A variable cathode resistor was used to obtain max gain it might have been at 80ohms? This is Pleiadss -1 or Pleiades 0? As heaters were disconnected while anode voltage still supplied the gain increased! It was such a discovery! So the heaters where fed by 5V from a 7805 regulator. It was also observed that on the AKG C12a the Nuvistor electron tube heaters are supplied by 5 volts instead of 6.3.
Back to the Pleiades breadboard jig. The above mentioned tubes were showed much much increased anode current when the high Megohm resistor was connected to Vb instead of cathode.
Good canidates were the 6SK7 and the EF183. In general good candidates for low voltage operation are varimu (variable gm) and high transconductance (gm) electron tubes.
It was listening experiments time and a Shure directional microphone was hung from the balcony facing towards walking people noise ambience etc.
An Altec green 47022 input transformer was capacitively coupled to the grid. And for output transformer a red Altec 15095 was connected directly to the anode as on fabulous 30's class A (electrons flowing all the time even through the transformer) circuits.
An micro ammeter was also connected through the anode circuit as was done in V1 electrometer pre preamp.
The Sony TC-D5 Pro must had been connected to the output transformer at its mic input. Monitoring was with the HD580 headphones.
Firstly the grid resistor was connected to ground. The sound was very faint.
When it was connecte to the anode it was a miracle! Everything could be heard clearly no matter how far away. It was big, dynamic and low noise. Undistorted.
The anode screen grid and suppressor grids were connected together as in the EMI RS61 front end. It was soon found that most of the work of attracting electrons is done by the screen grid which is closer to the electron could. It is most of the anode current which is typically 200-300 micro amps!
It has found by listening to the ambient noises through headphones that an even more natural sound is extracted when the grid resistor is connected to anode rather than Vb. So this is how it stayed connected.
Hliana was right, electrons have to be free.
After some time the Armstrong pair of 1917 was read or reread but now an explanation was found. That the grid impedes the flow of electrons. So there is a negative bias process inside the electron tube just by the boiling electrons leaving the cathode positive and the grid (negative) interaction.
So an external positive bias must be applied to compensate this effect and make very low anode voltage operation possible.
This V6 circuit was found to be very low noise.
And of course this amp started to be connected to the Grampian Gr1/L.
And since it was a joy to wind so few turns needed on the Magnetec nanocrystalinne cores an input transformer was made with only 3mH input inductance. And a 1:30 turn ratio?
This time the Grampian sounded so big without almost any hiss. And it sounded big as a classic electron tube mic but with smooth and detailed treble.
And suddenly the heater voltage wire was disconnected and the cathode temperature started to go down. And the sound and signal to noise ratio further increased! On this experiment the V6 was connected to the realistic disco mixer at its mic in.
This is how the V6 is born.
It was abandoned for some time since the V4 and V5 were made with similar principles. Also the V2 and V3 operating with the ECC82 with just 12V or even 6V at the anode. The V4 with 3.9V, EF183 and just 3.9 volts. The Pleiades V-1 was converted to having the grid biased from anode by 10Megohms. The other pre preamps use 4 to 8 Megohms as grid accelerating resistors.
The V5 is with 2 stages of EF183 but with no output transformer as the one stage V4. The V4 does not even have an input transformer. There is a 1/4in jack input so that inline input transformers can be connected to this hi Z input. It has to be noted that the anode to grid biasing creates an 100KOhm input grid impedance and because of low noise a very low equivalent input noise resistance, it typically may be 2.5Kohms.
It was time to revisit the V6 this time with a Pleiade input transformer of higher input inductance in order to try so many mics fr from eBay that were used for the V4, V5 tests. Also a Pleiades output transformer was wound too on a Magnetec 070 Nanoperm core.
And the V6 sounded so big, low noise but so much detailed for the MD21 and the MD211 which had just arrived. The MD211 sounded bass light compared to the MD21 which was fine connected to 140mH. The MD211 needs to see a higher input inductance.
At that point it was promised that no more mics would be bought after the MD211 and the MD214 lavalier. But after seeing and listening to the Grampian DP4 on Martin Mitchell's website it was forgotten, really!, and a Grampian DP4 was bought too. The low impedance DP4/L version.
The sound was jaw dropping. A great bass together with a great high frequency detail, smooth and edgy at the same time mid range. And no hiss. Ambient detail from almost everywhere as far.
So it seems it all started with the Grampian GR1/L trying to make it as quiet, low noise and as big and high frequency detailed as possible. One of the reasons the DP4 sounds so detailed is that the 100Kohm grid impedance devided by 10 devided by 10 (1:10 ratio input transformer) gives an 1Kohm load for a 25 ohm mic, 40 times more!. And the mic 25ohm output impedance is transferred to x10x10 or 2500ohms. But the V6 is so low noise having an equivalent input noise resistance of about this value! It is so low noise possibly because of the absence of any secondary electron emission effects due to the so small temperature and voltages involved. And from voice coil to grid there is only one input transformer with a primary winding of wire wrap thickness.
Signal Path, setup:
Grampian DP4/L - Pleiades V6 with Rag=8Megohm - Sony TC-D5 Pro - Sennheiser HD580
Thank you GR1/L. How can you be such an overlooked mic teaching so much and sounding so great too.
Thank you Grampian.
References:
Pleiades V6 Schematic
Neumann U47 schematic
On preserving transconductance with an anode voltage of 3V - euroelectron
Operating Features of the Audion - E.H. Armstrong
No comments:
Post a Comment