project 8

Page being initiated on: 6 September 2016


Status: 20 September 2016

revision: on 10 June 2022

Please notice my comments at the end of this web-page






YouTube first recording experiment film 0227


The foregoing chapters 5 and 6

We have reached a critical point, where so many different subjects being dealt with, that with the start of a new season we have a good opportunity so create a new webpage onto our ongoing Magnetophon Saga.

To refresh your memory a bit:

Early March 2013 we very kindly got donated a Telefunken T 8 type Magnetophon.

Including quite unique attached units, such as:

V 47/2 a genuine recording unit, including bias as well as erasing generator


V 41, the play-back amplifier; also designated, according the accompanying diagram, also occasionally used as a microphone amplifier. However, we employ it as to amplify the play-head signals. 

And we also got a set of two Neumann condenser microphones (Kondensator-Flaschen).

Our Belgian friend has apparently encountered many problems with the mounted capstan-motor.

First, according to the available documentation it should have looked differently and it was designed genuinely for a 76 cm/s tape speed. Whilst our friend operated a 38 cm tape speed. Viewing his capstan motor from outside, is seems to look like a wartime motor. Though, in those times, they operated a tape speed of 76 cm/s only; particularly in broadcasting environment.

In Belgium, I guess, they have modified the motor field numbers from 2 fields into 4 fields.

However, this motor behaved very poor and suffered apparently from 'hunting' effects.

Jaap Keijzer, our mechanical Guru, has tried everything as to get it work more or less appropriate. Result so far negative.

Early this year he started a new project modifying the motor mechanically. During his mechanical work he discovered, that the Belgians commenced really a rather poor job.

Normally capstan motors using plain-bearings (Gleitlager), at least for the upper bearing; instead, likely for having no better option, they have employed ball-bearings; with all the nuisance of (mechanical) 'rumble'.

What is a bit hampering is the fact that Jaap needs a rather long 'reamer' and such device isn't easy to obtain for a reasonable price.

Bernd Fischer from Germany gave me, very kindly, an AW 2 motor recently. Which is of an entirely different construction. Also the length of the capstan shaft doesn't fit to our situation; and additional nuisance, the motor diameter is also quite much bigger.

Jaap's first action was to make a Teflon like disk, so that the motor does not interfere with the T 8 mechanical situation just underneath the T 8 tape-deck.

However, a big nuisance was encountered, our AW 2 capstan shaft wasn't centred and showed a deviation of about 0.1 mm.

Thanks to Jaap's friend, the owner of a quite sophisticated mechanical tool shop company, they were ultimately able to grind and afterwards polish a new capstan shaft.

This was necessary anyway, because the AEG AW 2 Magnetophon was used for domestic purpose operated at tape-speeds of 38 and 19 cm/s (neglecting the very small margins after the comma).

Our T 8 machine as well as the AW 1 Magnetophon machines operated 9.72 mm capstan diameter; whilst the AW 2 machine used 5 mm instead.

Our first aim was to increase the capstan diameter again at 9.72 mm again. Combined with the AW 2 motor options we get:

operating the 4 field mode we get 750 rpm (3000 : 4 = 750; considering our European mains frequency of 50 Hz)

employing the 8 fields option, we get 3000 : 8 = 375 rpm.

We have got now two tape speed options 76 cm/s or 38 cm/s


The white coloured ring is visible between the top of the motor body and the tape-deck, allowing that the motor makes space for existing mechanics; as to fit onto our purpose


The AW 2 motor fits now quite well onto the situation underneath our T 8 Magnetophon deck


The reason we cannot operate a different capstan type is that it should exactly fit onto the diameter of the rubber pressure roller versus the capstan diameter


Viewing it just from a bit different angle

Please notice the brown-coloured just thicker roller diameter. This provision reduces considerably the friction between rubber roller and the capstan driving power. Because it only rubs just there where the tape is in between the roller and the capstan.

A great deal of luck is - that apparently since pre-wartime days the configuration of the 4 motor mountings screws haven't changed!



When you look carefully, you can notice that there exist a series with 4 and 8 field connections

Quite confusing was, that the shown numbers do not fit with the data provided on the AW 2 schematic. Secondly, in our configuration, the AW 2 capstan motor has to rotate into the opposite direction; viewing it from the top - our runs clockwise whereas the AW 2 operated anti-clockwise.


Our conclusion today, is, that this configuration does run fine. Albeit, that straight from the Telefunken T 8 Magnetophon concept of about 1948/49 they must have encountered tape running irregularities. There exists a photo where it can be seen that a complicated tape running provision being integrated between the left-hand side tape disk and the entrance of the head-arrangement. Whatever it might have once constituted, in my perception this was as to counter tape-feed irregularities! Which in some respect we still encounter, its appearance is like a kind of (small amount) flutter.

We did to day also our standard Lissajous tests, and these did not show different irregularities, than regularly encountered. This test is likely the hardest test possible, maybe too strict, but it provides a good understanding of what goes wrong. As the behaviour of the Lissajous tells us quite much about mechanical nuisances. In my perception much more that a regular "wow and flutter meter"; where we read off a level  given in a percentage.

I will ask Mr Engel, whether I may use this very informative photo.



On  7 September 2016

Today I have started my first experimental recording, using a Violin Concerto, because violins are very sensitive to irregular tape speeds.


Film 0227:    You are viewing our first experimental recording after the new adapted AW 2 capstan motor has been mounted underneath the T 8 Magnetophon deck. We encountered first some minor tape speed irregularities which minimized soon, origin apparently the left-hand side tape-reel motor; which is constantly acting as to keep stretched the tape running into head-assembly.     Up to 2/3 of the tape length all works quite sound. However, than increasing irregularities being encountered. The tape passing the capstan/rubber roller toward the right-hand side drum (Umlenkrolle), with combined spring-loaded lever, showing increasing amount of tape flutter. This is also visible in vibrating drum-arm. (Please forgive me, that I don't know the correct technical expressions to these devices)

My preliminary conclusion, the reason for the nuisance might be found in the right-hand side reel-motor. My first thought, is, that the motor-power isn't sufficient enough when the right-hand tape-reel gets increasing tape-reel diameter.

I therefore have to find out which resistor is responsible for the nuisance (the genuine schematic doesn't fit to our recorder version). This phenomenon can be repeated time and again and doesn't show a different behaviour.


Recording test setup



Making YouTube film recording


Viewing the running tape-deck during film recording

Please notice: the lever just adjacent to the drum-roller on the right-hand side is the one that later-on increasingly is showing tension (torgue) vibrations, when the right-hand tape is accumulating more than about 2/3 of the max tape length! But his phenomenon should have start before this figure, but slowly is increasing.



On 18 September 2016


I continued with determining what might cause the problems with the right-hand side tape-reel.


The trouble-causing right-hand side tape-reel

Particularly causing visible tape-running-vibrations (quite fast fluctuations) of the tape section leaving the capstan directing towards the right-hand tape-roller-drum and the attached lever; with increasing tape diameter the vibrations do increase as well.

My first thought, does the right-hand winding motor provide enough torque?


What I have measured first is: what voltage being supplied onto it in the various modes. Such as regular play/recording - winding right-wards and rewinding.

First play right-hand reel filled for 40 %   'U' (motor supply) ca. 110 V ac. This figure in contrast what is expected - changes very minor.

Test on request from Jaap Keijzer, of the Capstan motor supply voltage 'U' is 227 volt. Which being regarded as sufficient.

My first aim is concentrating upon the play-winding-up voltage supply of 110 V ac. Is this supply sufficient?


The very basic test principle

The only precaution to be noticed, is, that the Magnetophon start-up should be done in the mode where the T 8 machine feeds regularly the right-hand side winding-up motor, because one does not know what tuneable voltage should then be dealt with. 


For my own safety, I have mounted this provisional switch inside a just available cardboard box; its aim is to avoid hazard

The test result was on the one hand confirmed my suspicion that the irregular tape running being caused by some kind of insufficient driving power.

Though, another phenomenon came to light as well: the capstan motor when confronted with increased tape-torque started making additional mechanical noise (mechanical hum).

Apparently, signalling that its lock-onto-the-mains-frequency is not the correct move to approach the winding-up torque problems.


Shown is the current test setup

The 'variac' is standing on the ground. On the left-hand side is the digital meter with which the actual 'U' voltage being measured (controlled).


My instant thoughts currently - might the 'winding-up motor' causing the problems? If so, we might have been bothering with this phenomenon straight from the beginning in 2013!

We have such motor in spare, actually mounted within another T8 tape deck. But this one is missing the attached break-facility. For practical reason, I hesitate to demount our current matched break-system.

Therefore, I called up Jaap Keijzer whether he can support me with his 'mechanical brilliance'!


What next? 

AOB: Today, 10th June 2022, I, after long consideration, can look back at the fundamentals behind the 'curious hunting' aspects.

The 'manco' essentially is:

The rotation ratio between the driving mains frequency of 50 Hz is becoming critical when the rotation speed is becoming below a crucial rotation value.

This phenomenon occurs when we start with a full tape-reel on the left-hand side and its motor responding in a reverse movement as to keep some stress on the tape.

Similarly happens, when the right-hand tape-reel is becoming at a critical diameter, then a similar effect is occurring:

its rotation speed is too low, for the winding motor powered also from the same 50 Hz mains.

Why didn't mattered this phenomenon when the T8 machine had been developed (1944 - 1947)?

They operated then at a tape-speed of 76 cm/s; whereas we operate it at a tape speed of  (roughly) 38 cm per second; thus at a half speed!

What should we do to counter our problems?

Two options:

First: I suppose to replace the AW2 motor and mounting an AW1 motor instead, and operating it at 76 cm/s, again.

The essential down side, is: that our recording time being limited to 20 minutes and using a 1 km tape length!

Second:  We could / should commence some experiments.

I favour, at least, to supply the two winding motors with variable mains-frequencies. Particularly, supplying the 'hunting motor' with a lower supplying frequency, but maintaining a similar supplying voltage.

This could prove that my thoughts being, at least, valid.

However, I imagine that this could hardly being the solution, but at least proving what is causing our annoying problems.

Consequently - there hardly is a way around to re-establish the original tape-speed of 76 cm/s; by means of mounting an AW 1 motor again.




By Arthur O. Bauer