RekOKut Imperial II restoration project

Anything goes! Inventors! Artists! Cutting edge solutions to old problems. But also non-commercial usage of record cutting. Cost- effective, cost-ineffective, nutso, brilliant, terribly fabulous and sometimes fabulously terrible ideas.

Moderators: piaptk, tragwag, Steve E., Aussie0zborn

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

RekOKut Imperial II restoration project

Post: # 44635Unread post jesusfwrl
Sat Nov 05, 2016 8:44 am

IMG_1890_zpsm5mc6s4s (1).JPG
IMG_1891_zpsban8zogp.JPG
IMG_1902_zps7csnxetz.JPG
IMG_1893_zpscgyi4wur.JPG
IMG_1895_zpsebq1osds.JPG
IMG_1896_zpsrayyoaco.JPG
IMG_1898_zps8ucfhhfj.JPG
IMG_1880_zpsx3hysfgz.JPG
Since we don't have enough projects already, we thought we would start another one...

Strange things do happen, and one of them was being offered a complete RekOKut Imperial II package in astonishingly good condition, located within the EU, which theoretically means no customs complications and more reasonable shipping rates (in practise, we have noticed that for some of the last shipments we did within the EU, to and from specific countries, some of the old fashioned customs bureaucracy has suddenly been re-instated, after many years of absence. They don't charge import duties within the EU yet, but it is a pretty clear signal as to the future of the EU, beautifully demonstrating how the world keeps on turning much like a record and nothing remains the same for too long).

We just couldn't refuse. USA-made vintage recording equipment, in usable condition, do not pop up around Europe every day...

The package consisted of:
1. TR43H, 3 speed turntable with a 12" platter, and the synchronous hysteresis AC motor fitted, rather than the induction motor found on other units.
This is 60Hz of course, but we have already solved such minor compatibility issues when we got our Fairchild lathe a while back. We have 115VAC, 230VAC and 380VAC, 50 and 60Hz, in single phase, two phase and three phase configurations, permanently installed in the studio.

2. Master Pro Model M12S overhead recording mechanism

3. Audax model RH5 monophonic magnetic cutterhead.

4. Tonearm, this looks like the S120, but it seems to be fitted with S220 counterweight. The headshell connector has 4 pins but the arm is wired for mono at present.

5. General Electric VR-II cartridge. This cartridge features a rotating mechanism with 2 styli, one for coarse groove and one for microgroove recordings.

6. R8B cutting amplifier. This is an interesting hybrid, featuring a transistorised pre-amp stage and a vacuum tube power amp stage.

7. Several boxes of unused lacquer, some even sealed!

8. Two cutting styli which appear to be made from diamond. They do not look like the work of Steven (Myshank), or any other stylus I have ever come across. Detailed pictures of that will follow at some point.

We were expecting to have a lot of work to do since none of this was really tested and the condition was unknown. Luckily, it mainly suffers from lack of use,as it had been sitting somewhere collecting dust for several decades. Once most of the dust was removed, most of the parts appear to be in fantastic condition but in need of a full restoration to remove the grime that has accumulated in all the moving parts.

The machine had spent most of its life in the USA before being imported to Europe along with a lot of other audio equipment, not related to disk recording.
There is some evidence of previous ownership by a company called "Galas Unlimited" in the USA, which I know nothing about.
IMG_1877_zpsbpse0aiz.JPG
IMG_1878_zps2uibrxzx.JPG
Before plugging anything in to a power source or trying out if anything works, ancient machines must be carefully inspected, cleaned and lubricated, to prevent damage. For instance, there was no oil in the centre bearing, but plenty of grime instead. The centre bearing shaft on this model is extremely thick and sturdy. It appears to have survived any attempt to see if this thing works by previous owners.
IMG_1882_zpsewdsbd3o.JPG
The rubber idlers appear to be in pretty good shape. The one mostly used was the 78 rpm, as evidenced by increased wear in its centre bronze bush. The 33 and 45rpm idlers seem practically unused. It was evident that the machine had not been lubricated for several decades. Luckily, it was also not really used during that time either. The power switch for the main motor is integrated in the mechanism for engaging the idlers. This switch was damaged, enabling only the 78rpm speed to be engaged after considerable effort, which probably explains the wear on the 78rpm idler, if used without lubrication. I originally thought I would need to replace the switch, but I just opened it up and repaired it for now.
IMG_1871_zpsue4rqgjm.JPG
IMG_1874_zpsxkvr8jzg.JPG
IMG_1876_zpsx2vxy8xa.JPG
Slowly stripping parts away to clean, inspect and lubricate. The idlers had a bit of old oil splashed on them, which was cleaned with warm soapy water. The platter rim was also thoroughly cleaned, along with the centre bearing, idler spindles, and all the individual parts that make up the overhead mechanism.
IMG_1973_zpsp9iswelw.JPG
IMG_1975_zpsxnrrx62p.JPG
IMG_1976_zpssodo2vhj.JPG
IMG_1974_zpspuswaepm.JPG
IMG_1977_zpsc5abolls.JPG
IMG_1978_zpsryrp5bdl.JPG
Very simple mechanism. Plenty of room for improvement as well, although we will try to keep modifications to a minimum, and implement them as inconspicuously as possible, in an effort to maintain the original perspective of the machine, since it is in such good state.
IMG_1979_zpsxrzmunzw.JPG
IMG_1980_zpsxhfslpps.JPG
The height adjusting mechanism is the only part which does not seem to be functional. The threaded part is meant to rotate on the shaft, prevent it from an uncontrolled movement through friction provided by the spring below it. When the height is correctly adjusted, the lockscrew on the side is tightened and the overhead mechanism is meant to be rotated into position though the rotation of the shaft within the threaded height adjuster which is now firmly locked in place.
This part appears to be stuck.
I would like to hear from other Rekokut users regarding this part: Does yours function as I described above, or are these parts known to get stuck after a while?

I initially thought that the spring tension might be incorrectly adjusted, so I removed the spring entirely and the threaded part still will not budge. It will do, with considerable force, but not under normal operating conditions. I will most likely use a bearing puller to remove it from the shaft and inspect and cure the problem accordingly. I suspect the problem is due to corrosion, but we will soon find out.
IMG_1966_zpskvatwf6a.JPG
IMG_1971_zpsxzd2utkd.JPG
IMG_1911-s_zps964dnb1m.JPG
IMG_1907-s_zps5p5hvv0k.JPG
IMG_1908-s_zps144xtkbv.JPG
IMG_1964_zps9mvrad9r.JPG
It glows!
The ventilation grills of the cabinet had come loose and all the steel nails were lodged in strategic places in the circuit, ensuring sparks, fire, and other special effects for the one who would carelessly plug it in. One more reason to never just plug in old equipment to see if it works without a thorough inspection.

I opened it up and removed the offending visitors and then politely started waking it up on the electronics bench with a variac.
I started with around 10VAC and within a few hours had reached the full 115VAC, with no special effects. The secondary of the output transformer (the output connection for the cutterhead or speaker) was connected to the appropriate dummy load. You definitely do not want to be plugging in an amplifier of unknown condition for the first time after years of not being used, with a cutterhead connected to it!
After several hours of testing, the amplifier seems to be performing great on the bench. Audio test to follow shortly.

The pre-amplifier section used the RCA 2N406 germanium transistors, the power supply uses RCA 1N1764 silicon diodes in a bridge rectifier configuration, and the vacuum tube complement features a Sylvania 6AN8 and a pair of Sylvania 6973.

The components used in the circuit seem to be of pretty good quality, which explains why they still function nowadays. I am not a big fan of re-capping things just for the fun of it, I have plenty of 30's and 40's circuits still operating on the original capacitors with no issues.
IMG_1965_zpsoyvnd6c6.JPG
This is what a variac looks like. If you are into waking up long forgotten equipment, get one of these.
IMG_1962_zpstbkrywzr.JPG
Finally, the turntable was put back together and carefully tested. The run-out, as measured when rotating the platter by hand, was +/-15um, which is a lot better than I was expecting. I used SAE10 motor oil for the centre spindle, which seems to be working fine for now, with the platter spinning happily and being super quiet. However, given the design of the bearing, it could be that it would be best to use a more refined hydraulic oil such as to Mobil DTE, which is used on a variety of precision machine tools and a lot of people use on their Neumann lathes. I have plenty of it, so I might try it out and see what performs best.
IMG_1884_zps7alfjteq.JPG
Pretty good looking!
IMG_1959_zps3qge1mee.JPG
Speed stability seems excellent. Confirmed with strobo disk. The motor and idler drive system appears to be sturdy enough that even trying to slow down the platter by hand has little effect on rotational speed. This is definitely a well engineered turntable, well suited to cutting disks, keeping in mind of course the inherent limitation of an idler drive transport and platter driven carriage.

Any input on the hight adjuster issue would be much appreciated. I am very curious if other have experienced the same problem.

More to follow...
You do not have the required permissions to view the files attached to this post.
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
mratx
Posts: 121
Joined: Sat Aug 01, 2009 6:54 pm
Location: Pittsburgh, PA, USA

Re: RekOKut Imperial II restoration project

Post: # 44641Unread post mratx
Sat Nov 05, 2016 3:31 pm

I'm not sure that I follow the issue with your height adjuster. The silver ring should adjust up and down the threaded shaft, the threaded shaft does not rotate on it's own, only as part of the mechanism it is attached to, which rotates where it connects to the one end piece of the overhead (the ballhead type of bearing) . On mine, I needed to remove the hex head set screw that goes down through the end piece of the overhead and clean things (I used brake cleaner) because there was grime and dried crud. The hex set screw is tightened to adjust the rotation of the bearing below it, if it's too tight the mechanism will not rotate.

Mark

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44643Unread post jesusfwrl
Sat Nov 05, 2016 4:22 pm

If the threaded part with the flat is not meant to rotate on its shaft, then how would you rotate the entire overhead from its stand to the centre of the platter?

If the lockscrew is tightened against the flat, then the overhead cannot be rotated unless the threaded part can. Or, to put it more precisely, once the threaded part is locked with the lockscrew against its flat, the only way to rotate the entire overhead is if the shaft would rotate within the stationary threaded part.

If it is not meant to rotate, then why does it have spring tension under it?

If I understand your post correctly, you are referring to the ball joint, with the lockscrew accessible on the top of the overhead, under a lid. This part is fine on my lathe.
You can rotate the overhead from there technically, but there are two considerations about that:

1. Shouldn't the pivoting action of the overhead vertical pivot be along the length of the overhead mechanism?

2. Isn't the whole point of lockable adjusters to set it up and lock it so you don't have to keep on readjusting it all the time?

This all points to the threaded height adjuster as needing to be rotatable on the shaft against spring tension to prevent wobble. I think this is how it was originally meant to work at least, no clear mention of any of this in the associated literature.

How do you rotate your overhead to move it to the stand to its operating position?
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
mratx
Posts: 121
Joined: Sat Aug 01, 2009 6:54 pm
Location: Pittsburgh, PA, USA

Re: RekOKut Imperial II restoration project

Post: # 44647Unread post mratx
Sat Nov 05, 2016 6:03 pm

I don't believe that the threaded section is supposed to rotate in normal operation. On mine, it's set so that the pivot above it is set as parallel to the leadscrew when rotated into place, I loosen and tighten the hexhead lockscrew to swing the arm as listed in the manual. The nut on my threaded section is held in place by a cotter pin, and the threaded section does not rotate, my assumption is that it was set at the factory and is not meant to be adjusted. At least I never have, and none of the literature I have says anything about doing so. I'm only a dabbler, not a professional, so I may be missing something, but this is the second m12 I've had and the other was the same.


Mark

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44651Unread post jesusfwrl
Mon Nov 07, 2016 4:27 am

Thanks for your reply. I have had another look at the manual and the overhead mechanism.
Among other things the manual also clearly states that "the Imperial II is the worlds finest disk recorder". :lol:

I am posting the drawing from the manual with the names of the parts so we can eliminate uncertainty:
overheads-ROK_zpsentfvcd5.PNG
So, the mount flange (1) is securely bolted on the turntable plinth. The drive flange (18) is resting on the stand, located on the front corner of the plinth.
The swivel (4) is oriented as shown on the drawing, with the overhead in this position. The elevating lockscrew (2) is tightened. The lockscrew located under the lockscrew cap (7) is also tightened. In this setting, I can lift the lathe handle (27) upwards and the entire mechanism will be tilted with the lathe handle moving upwards and the crank (9) pointing downwards. This movement is allowed by the swivel (4). However, in this setting I cannot rotate the overhead mechanism towards the platter to place it in operating position. To achieve that, there are three options:

1. Unscrewing the elevating lockscrew (2) to allow rotating of the elevating screw in the mount flange (1). There are two problems with this approach: when the overhead is rotated into operating position, the flat on the elevating screw is no longer aligned with the elevating lockscrew (2). As such, it would not be wise to attempt to tighten the elevating lockscrew against the threads. Without tightening the elevating lockscrew there is considerable play between the elevating screw and the mount flange (1). The other issue is that the elevating wheel (3)remains stationary on the mount flange (1)due to friction, so rotating the elevating screw while keeping the elevating wheel stationary would upset the height adjustment setting.

2. If we are to leave the elevating lockscrew (2) tightened against the flat, to not upset the height adjustment and not introduce wobble, we could instead unscrew the lockscrew under the lockscrew cap (7) and rotate the overhead mechanism from the ball joint between the swivel (4) top and the casting. This, however, would upset the normally orthogonal relationship between the swivel shaft and the slide bar (16). The swivel (4), would no longer serve as the pivot to tilt the entire overhead mechanism upwards. Instead, it would allow the mechanism to be tilted forwards or backwards while in the operating position and it would also interfere with carriage movement to the extreme right.

3. The only other way in which it would be possible to move the overhead mechanism from its rest position in the operating position without unscrewing the elevating lockscrew (2) or the lockscrew under the lockscrew cap (7), would be if the shaft running through the hallow elevating screw, pinned to the swivel casting on one end, and having threads on the other end for the nut which adjusts the spring tension against the elevating screw, would be able to rotate within the stationary elevating screw which is kept locked by the elevating lockscrew (2). In this case, all the adjustments could be locked and the overhead mechanism could still be rotated between rest position and operating position without any need for re-adjustment.
This appears the be the intended method of operation, taking a wild guess from the designers perspective. If this would not be the intended method of operation, there would be no justification for the use of a threaded sleeve on a shaft held under spring tension, utilising a spring, three washers, a nut of special construction with milled slots, a split pin, a polished shaft with a threaded end, and a hole drilled into the threaded end for the split pin to securely lock the position of the nut, thereby ensuring constant calibrated spring tension. If there was no intention for rotation in this part, it could have also been constructed as a solid threaded rod, with a flat ground on one side. No need for springs, washers, nuts, complicated machining, and someone's time in assembling and adjusting this at the factory before shipping out to customers.

There is no other obvious reason to implement these parts in this configuration. This configuration greatly increases the manufacturing cost of the unit compared to just a solid single part. I don't think they would be doing this without a good reason.

If this part is stuck on many of the M12 overheads out there, it could be a bad design of the parts from the beginning, or less than precise manufacturing, or a not so careful choice of materials for each part. In my opinion, differential corrosion is very likely to have caused them to stick together.

So, which one of the three methods do you currently use to move your overhead from rest position to operating position?

I should also note that in the manual for M-12S, under "C. Operation" there is no clear instruction to loosen or tighten any lockscrews to swing the lathe mechanism from rest position to recording position and back.
You do not have the required permissions to view the files attached to this post.
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
mratx
Posts: 121
Joined: Sat Aug 01, 2009 6:54 pm
Location: Pittsburgh, PA, USA

Re: RekOKut Imperial II restoration project

Post: # 44656Unread post mratx
Mon Nov 07, 2016 12:04 pm

I use a variation of method 2. I loosen swivel lockscrew #7 enough to rotate the arm, and elevating lockscrew #2 enough to use the elevating wheel #3. I then move the arm into place, adjust the elevating wheel to ensure the arm is parallel to the disk, and then tighten lockscrew #2. I then start the turntable, making sure the drive flange #18 is fully down and level and tighten the swivel lockscrew #7.

Mark

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44756Unread post jesusfwrl
Sun Nov 13, 2016 2:34 pm

The main problem with this machine was endless amounts of dirt and grime having found their way into all the places where they shouldn't be.
Had it been used like that, it would have been totally destroyed within a very short period of time.

The grease had turned into a solid soap-like abrasive compound and there were no traces of oil on parts that should be coated with a film of oil. Essentially, the cure is to strip the machine down completely, clean each part individually, lubricate and reassemble.
DSC02462-s_zpsg4xnjnm0.JPG
The worm gear is the part that goes over the centre spindle and transmits motion to the leadscrew. This has a seal on it which is supposed to catch any dripping oil. This was completely disintegrated, with little hairy bits spreading all over the place, so I fitted an absorbing seal that I found in my pile of interesting bits. No idea what it is made of, but it does the trick pretty well. I find that as the size of my collection of interesting bits increases, so does my ability to solve problems like that in a timely fashion.
IMG_2044_zpsn940t4ku.JPG
The leadscrew was so filthy it took almost 2 hours of careful cleaning to get it to a respectable state.

As for the height adjuster, my assumptions have proven correct. The threaded sleeve is indeed meant to rotate upon the shaft, but it was prevented from doing so due to differential corrosion. The threaded sleeve had to be removed with the aid of a bearing puller and treated to remove the built up of the corroded nickel plating. Following this, it was reassembled and the spring tension was adjusted appropriately.
DSC02459-s_zpsqfpg25h5.JPG
DSC02451-s_zps6ql5etc6.JPG
DSC02454-s_zpsw0pyg3jm.JPG
Without the spring installed, the sleeve will now easily spin upon the shaft, as demonstrated in the following short video:
The overhead lathe mechanism can now be rotated from its rest position to its operating position on the centre spindle of the turntable platter, without having to loosen either of the two lockscrews. Therefore, once the height and geometry is correctly adjusted for a certain type of blank disk, these adjustments can be locked by means of the two lockscrews provided, one engaging on the flat of the threaded sleeve of the height adjuster, and the other one engaging on the top of the ball joint, hiding under a cap on the hand crank end of the assembly. Thus, as long as the same type of blank disk is used, the adjustments shall remain locked and no further adjustment shall be necessary each time the machine is used. This makes things a lot easier and safes a lot of time and setting up the recording process.
IMG_2050_zpsliiqwhww.JPG
IMG_2052_zpsgjtnvpki.JPG
This is how the alignment of these parts should look like when correctly assembled and functional. The flat on the threaded sleeve of the height adjusting mechanism should remain aligned with the lockscrew at all times, both in operating position as well as in rest position. The swivel for lifting the assembly upwards should form a 90 degree angle to the leadscrew, again at all times and in all positions. When everything is adjusted correctly, there should be no discernible play in any of these parts. There should also be no discernible play on the suspension pivot bearings. This particular assembly had plenty of play everywhere, but after the complete strip down and reassembly, it was adjusted carefully to take up play, but without over tightening which could prevent the smooth operation of the moving parts.

I highly recommend sorting the height adjuster sleeve out for users of this overhead assembly, it makes life much easier.

For the not technically minded, I would be happy to offer this as a service. Please feel free to contact me if you would like your Rek O Kut repaired or restored.

More soon regarding the amplifier and head.
You do not have the required permissions to view the files attached to this post.
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
grooveguy
Posts: 425
Joined: Thu Jun 22, 2006 5:49 pm
Location: Brea, California (a few miles from Disneyland)
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44760Unread post grooveguy
Sun Nov 13, 2016 3:24 pm

Fantastic, Jesus!

What a patient and accomplished restorer you are indeed. Yours must be a later issue, what with miniature tubes and even transistors(!) in the amplifier.

I have always admired Rek-O-Kut machinery, and obtained an M12S like yours (but no amplifier) back in the 1970s, fitted with a Presto 1D head. Used it for years that way, and then 'upgraded' it with the beefier R-O-K 16-inch overhead mechanism and a Fairchild 'Olson' cutter:
Stock shot.jpg
It continues to work for me very well. I have great respect for that turntable, much quieter than R-O-K's playback turntables of the same era. In his past, a friend of mine had your same lathe with the Audak head you have, this must have been in the early 1960s. He took it to a local ballroom and made some 'direct-to-disc' dance band recordings that were superb.

Do keep us updated with your progress.
You do not have the required permissions to view the files attached to this post.

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44846Unread post jesusfwrl
Sat Nov 19, 2016 2:17 pm

Many thanks grooveguy.

Your RekOKut setup looks absolutely stunning!
I love the Fairchild cutterhead.

I would assume that our Imperial II system dates around the mid 50's, but this is just a guess.
Location recording is definitely on our to do list with the RekOKut in the near future.
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44862Unread post jesusfwrl
Mon Nov 21, 2016 1:07 pm

Spiros did a great job restoring the cabinet of the R8B amplifier. He applied rust converter to the rusty spots, removed the perforated ventilation panels, stripped them from the old flaking paint, and had them powder coated, and repaired the wooden structure. It now looks like it just came out of the factory, which it did...but a different factory...Well, not really a factory, a workshop. Given how we do things at Magnetovolt, we shouldn't really call it a factory, since this implies mass production, which we do not do.


Image

Image

Image

Image

Image

Image
Image
Image

Image

Image
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
grooveguy
Posts: 425
Joined: Thu Jun 22, 2006 5:49 pm
Location: Brea, California (a few miles from Disneyland)
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44866Unread post grooveguy
Mon Nov 21, 2016 3:03 pm

Quick question: After removing the rust from case hardware, did you re-plate the pieces with anything? I know there are some brush-applied electroplating kits for model railroad and car restorer types, but don't know if you went that far.

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44941Unread post jesusfwrl
Sat Nov 26, 2016 7:00 am

Grooveguy, I have no idea. I have sent the link to this thread to Spiros, who is not yet active on lathetrolls, but I know he is reading... He is the most qualified person to answer your question.

In the meantime, I finally found the time to do some quick tests on the R8B amplifier.

Image

First and foremost, checking the current through the tubes, which is easiest done using suitable test probes between the tubes and their sockets. However, the 6973 tubes used at the output stage have very different pin outs to other noval based tubes that I already have aesthetically pleasing probes for, so we just put together these monstrosities which do the job just as well, if you are prepared to have your visual sense offended in the process. The tubes were found to be completely out of whack, way too much current going through both of them, far exceeding the maximum plate dissipation. This was promptly rectified by adjusting the bias voltage, since this amplifier makes use of a separate negative voltage supply and operates the output tubes using fixed bias, with grounded cathodes.
Luckily, I did not operate the amplifier in this state for long, although it appears that previous owners might have done so, since the two output tubes show very different characteristics to each other. All the tubes are still the original Sylvania USA tubes that the amplifier most likely left the factory with. A quick search showed that for some odd reason the 6973 tubes are still being manufactured by some Russian factories, so at some point we might fit some new ones on there.
I then proceeded with adjusting the AC balance of the output tubes using the control provided, for minimum distortion(maximum cancellation of second harmonic distortion due to push-pull action can only be achieved if the tubes are correctly matched/balanced). These are still far off from my usual standards, but it is the best I can do without replacing tubes or modifying the amplifier, so I proceeded with the test.

Image

The amplifier performed better than I expected in most respects, apart from output power. The maximum undistorted output power on a sine wave was only 14W, rather than the optimistic specification of 20W. I suspect that the 20W was a theoretically predicted value during the circuit design stage, without taking into account the negative feedback. The amplifier would definitely put out 20W or even more, but it will be heavily distorted. This is measured into an 8 ohm resistive load as with all the measurements taken. The limit does not appear to be in the preamplifier section, the output of which remains quite clean even at much greater amplification. The limit seems to be imposed by the output stage itself.

The measurements were all taken by connecting a signal generator with a 600ohm output to the tuner input of the R8B. The amplifier maintains correct polarity throughout and the meter is calibrated so that a 40% reading is reached at a 0.56W output (0.265A). 80% = 2.25W(0.53A) and 100% = 4W(0.707A). Interestingly the amplifier output has a flat frequency characteristic, and does not include a RIAA pre-emphasis curve in recording mode!
The frequency response was indeed within the -+2dB from 20Hz to 20kHz, with +2dB reached at 30Hz, reducing back to 0dB at 10Hz, but heavily distorted as expected at such a low frequency. The HF response drops to -2dB from around 15kHz onwards and reaches -3dB at 22kHz. The -6dB point is at 105kHz which gives you an idea of what will arrive at your cutterhead coil of you continue boosting indefinitely past the 75us time constant of the RIAA pre-emphasis curve. The Audax RH5 head is meant to not be able to go past 13kHz, mechanically, but extremely high levels of HF content would still run through the coil, being dissipated as heat. There is a resonance at HF at around 90kHz, where the signal suddenly rises to almost 0dB and then sharply back down to -6dB at 105kHz.

The phase response of the power amplifier section is pretty much spot-on due to the use of overall negative feedback. However, the negative feedback loop, enclosing the output transformer, only goes back to the pentode input stage, not including the transistorised preamplifier stage. This creates an overall phase deviation of around 10 degrees at 1kHz between input and output, which stays pretty much the same throughout the 20Hz - 20kHz range. The phase shift reaches 90 degrees at 45kHz at HF and 8Hz at LF. At the 90kHz resonance the phase shift is 180 degrees, possibly explaining the resonance as a positive feedback situation, which however, remains very well under control. The stability of the amplifier is excellent.

The voltage gain from the tuner input to the output is 35.5dB. The preamplifier alone can offer significantly more amplification, but due to clipping of the output stage, I only tested it up to 14dB. The signal input for the tests was 0.18Vrms, which is quite low by modern standards. This, however, makes this amplifier ideally suited for use with one of the passive RIAA pre-emphasis circuits we have constructed based on the relevant AES papers by Stanley Lipschitz.

The loss through the passive RIAA pre-emphasis network would bring normal signal levels to the correct range for use with this amplifier. Even more interestingly, the phono input of the amplifier contains a RIAA de-emphasis network!

I was very intrigued by the choice of the 6AN8 in this design and investigated further hoping I might find a hidden automatic gain control circuit, since the pentode section of the 6AN8 was originally intended for such applications. I drew the schematic by looking at the unit and found no hint of an AGC circuit incorporated into this unit. Apart maybe from some limiting action imposed by the choice of operating points on the individual stages. Nothing too fancy though...
Here is a simplified drawing of the circuit for those of you interested.

Image

The preamplifier section can be bypassed altogether and the audio connected directly to the pentode section of the 6AN8, offering a bit extra headroom and possibly a bit less noise, but some modifications would be required for this. So, the pentode section of the 6AN8 functions as a single ended amplification stage direct coupled to the triode section which acts as a split load phase inverter, driving the 6973 output beam power tubes in push-pull, class AB1.
Typical "pentode" connection is used for the pair of 6973 tubes.

The hum at the output was measured at -63.5dB, and other noise at -74dB. These are unweighed measurements. The only thing left now is to inspect the cutterhead and do some test cuts!
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
grooveguy
Posts: 425
Joined: Thu Jun 22, 2006 5:49 pm
Location: Brea, California (a few miles from Disneyland)
Contact:

Re: RekOKut Imperial II restoration project

Post: # 44945Unread post grooveguy
Sat Nov 26, 2016 11:31 am

Wow, Jesus,

That is a most comprehensive testing and evaluation of that amplifier. Indeed, the performance is better than what I would have expected too. So what did Rek-O-Kut, themselves, do for EQ, anything at all? My friend who used one of these recorders for live dance-band sessions had remarkable results, with the discs sounding very much like the live performance. He was using Sony condenser mics, if I recall, but no additional EQ past whatever was in the ROK.

Your mention of the amplifier bandwidth and what might befall the cutterhead is a very good point. I don't know how many Presto 1D heads were destroyed by engineers rewinding masters with the tape recorder head-gate closed. Doubtless Stan's 'reverse RIAA' circuits are within a millibel of what they should be, perhaps way past 20kHz. Summing the output of a low-Q peaking circuit with the same signal at its input will give a nice 75us lift right to the peak, and then it will fall off quickly after that. Something we do in FM broadcast audio processors to keep the 15kHz low-pass filter from having to work too hard.

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 45015Unread post jesusfwrl
Thu Dec 01, 2016 11:52 am

I was also wondering what the original intention was regarding pre-emphasis. Since the turntable also includes a 78 rpm speed, which was probably still widely used at the time, a RIAA was not necessarily the desired pre-emphasis equalisation curve. So, it could be that the lack of any pre-emphasis circuit was intended to maintain compatibility with different curves through the use of external networks.

I remember reading somewhere that the Audax RH5 head was meant to have a built-in pre-emphasis...I don't remember where I read this, but it seems highly unlikely. The physical size of the head would make it too small to contain a passive electrical network in addition to the internal workings, and a mechanically implemented pre-emphasis with three precisely defined time constants sounds like something straight out of a science fiction film. Not to mention that even if someone would have managed to design this (either through intense mathematical calculations or through an insane amount of trial and error) its performance would greatly depend on the environmental conditions and would drift with age. I did not yet have time to inspect the head or cut with it, so we shall be finding this out in due course. The manual simply mentions that the bass and treble controls are provided for the desired recording and playback equalisation, whatever that means. The time constants of the controls though do not match time time constants of any commonly used emphasis curve.

As for pre-emphasis implementation, the RIAA curve is not defined past 20kHz. The difference I think is that in FM broadcasting, the medium is very specifically defined as being band limited, whereas the disk medium is not an intentionally band limited medium. In the early days it was limited by the development of the equipment and later on, when the equipment became capable of exceeding the 20kHz mark, each manufacturer of recording or reproducing equipment seems to have gone their own way in imagining how the extended range capability should be implemented.

In theory, the pre-emphasis and de-emphasis curves should complement each other. In practise, manufacturers have always used different pre-emphasis curves and passbands.

In the days before super wide range cutterheads, the only consideration was not overloading the cutterhead, but wide range amplifiers were also not that common anyway. But, later on, another important consideration was whether the reproducing stylus would be able to trace the exceptionally short wave lengths that could be recorded on disk. It is a very interesting question whether there should be any standardisation of the additional time constants required for the practical implementation of the RIAA pre-emphasis equalisation, and if so, what should their value be? There has been some discussion lately, or rather speculation, on what time constants were implemented by different manufactures of disk mastering equipment, by individuals interested in the accurate reproduction, including manufacturers of phono stages. I have read through all this with interest. I have yet to come across a complete and convincing approach on this subject.

Your approach in the broadcasting equipment sounds like a really good idea.
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
mratx
Posts: 121
Joined: Sat Aug 01, 2009 6:54 pm
Location: Pittsburgh, PA, USA

Re: RekOKut Imperial II restoration project

Post: # 45019Unread post mratx
Thu Dec 01, 2016 3:28 pm

Have you considered replacing the carbon resistors in the audio path of the amplifier with metal film (or even better metal foil) ones to reduce resistor noise? Or, if you want to go all out, changing the pots to conductive plastic element ones? I personally just swapped in a modern SS amp for my tube one instead, but if you want the tube sound with a little less noise on older equipment, some resistor swapping may help. Unless you consider the resistor noise part of the unique sound (as in guitar amps.)


Mark

User avatar
socialroots
Posts: 161
Joined: Tue Nov 11, 2014 2:33 pm
Location: Dulan village,Taitung,Taiwan

Re: RekOKut Imperial II restoration project

Post: # 45022Unread post socialroots
Thu Dec 01, 2016 9:16 pm

Jesus , just want you to know that as a Rek o Kut owner I am following your recent this restoration of yours with great pleasure as your journey is the most detailed info ive come across thus far about rek o kut cutters and im sure all your ivestigating and probing will help out all rek o kut owners.
Does anyone have any info or know the difference between the Audax rh5 and the Audax h4?
Seems to be that all the Audax shiny chrome Audax heads I see are rh5 but I have one that looks identical to all those rh5 , but its says h4 on it.
Maximum respect and a big shout out from Taiwan,
Patrick

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 45027Unread post jesusfwrl
Fri Dec 02, 2016 5:21 am

Mratx, yes, I have considered replacing components, but I am a bit divided on the matter, leaning towards keeping it as it is, since this amplifier is more of a museum item to me rather than a high performance amplifier. Since we found this in such good condition, we might as well just keep it as it is until something really needs replacement.

Regarding resistors, I am a big fan of metal film and metal foil resistors and use them extensively in my own designs. However, for situations were there will be a significant DC potential across the resistor (plate load resistors), I have found it preferable to use non-inductive wire wound resistors. Prototype testing usually influences my final decision, since the difference between components and circuit layout is usually easily measurable with test equipment. The listening tests usually correlate well with the scientific measurements and I use both approaches for all the equipment I design. When it comes to modifying existing equipment, it is a bit of a tough choice, as in the Rek-o-Kut I was also tempted to replace the electrolytic capacitors with other types, which would probably not fit in the chassis, replacing the pentode input stage with a triode, adding dedicated drivers between the phase splitter and output tubes, rebuilding the bias and balance circuit to a better design, replacing the solid state preamplifier section with a vacuum tube one of my own design, and by the time I do all this it is no longer a Rek-o-Kut.
So I might as well keep it as it is and design a better one from scratch. If ultimate performance is desired, a new design using new high quality components, and possibly specially selected NOS vacuum tubes, would effortlessly outperform any of the vintage tube amplifiers out there. So by now, I am a bit sceptical about modifying vintage gear, often preferring to just design them from scratch with ultimate performance in mind from the beginning. I would also do away with the pcb approach and use point-to-point wiring, with solid silver wiring. But then again, the cost of building such a high end amplifier would be significantly higher than the cost of buying a vintage amp and just swapping out some resistors. The performance you can expect varies accordingly. In the end of the day, if the goal is to drive a vintage magnetic monophonic cutterhead, this will be your biggest performance limit anyway.

Image

This is what our latest amplifier design looks like. This was made as an instrument amplifier head, but it far outperforms the Rek-o-Kut amplifier, so it could easily be used to drive a magnetic cutterhead. Although the power output is a bit overkill, since this beast can put out a continuous 200W sine wave into 8ohm. It uses tube rectifiers as well, and absolutely no electrolytic capacitors in the entire amp, also no carbon resistors. You can expect excellent dynamics and an instantly fried cutterhead coil if you are not careful.

Patrick, many thanks for your kind words.
I do not have much info (yet) on the various Audax cutterhead models. I am planning on removing the cover on mine in the very near future and posting some detailed pictures of what's inside. If you do the same we could compare and see if there are any obvious differences.
What are you using for the RIAA pre-emphasis?
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

User avatar
SpyrosMV
Posts: 4
Joined: Tue Nov 29, 2016 2:23 pm
Location: U rope

Re: RekOKut Imperial II restoration project

Post: # 45053Unread post SpyrosMV
Sun Dec 04, 2016 2:58 pm

grooveguy wrote:Quick question: After removing the rust from case hardware, did you re-plate the pieces with anything? I know there are some brush-applied electroplating kits for model railroad and car restorer types, but don't know if you went that far.
Hi groove guy,

I did not go that far indeed. I used a rust removing solvent, which did the removing job and afterwards i just let the surfaces.. "breathe". Having tested this method of rust removal, i have found the result long lasting, even when the parts are left exposed in considerably high humidity conditions. The Rek-O-Kut is gonna be stored and used in quite good temperature and humidity conditions, so i didn't find the need to go that far and trouble with more work on it at the moment. If for some weird reason it will turn out oxidizing in the future, i will consider some alteration of this kind. For now it looks as a really well maintained, original lathe cutting equipment of the 50's, and i d like to keep it as close to this as possible. It's beautiful!

Your Rek-O-Kut looks also really beautiful and in excellent condition. I was wondering about the white color of it. Was it manufactured like this or has it been painted afterwards?
http://www.magnetovolt.com
Exceptional Quality Custom Audio Equipment

User avatar
grooveguy
Posts: 425
Joined: Thu Jun 22, 2006 5:49 pm
Location: Brea, California (a few miles from Disneyland)
Contact:

Re: RekOKut Imperial II restoration project

Post: # 45054Unread post grooveguy
Sun Dec 04, 2016 3:43 pm

Hi, Spyros,

Nice job on that case! The hinge looked so nice I figured you might have replated it. I've often been tempted to get one of those little battery-and-a-brush plating kits that are popular among model railroaders. I don't think the plating from those is ever very thick, although I suppose you can just keep on going over and over it. But having good resistance to the weather, even if indoors, might be worth the trouble. Anyway, the part would not have to be removed and immersed, nice in a case like this.

Thanks for the comment on my Rek-O-Kut, which is actually two different models: the turntable from the 12" Imperial series, and the M5S overhead that originally would have come with a 16" turntable. I stripped-down both parts completely, removed all existing wrinkle-finish paint, and had those parts painted by the shop that does front panels and other parts for my business. That color is, in fact, the color that our panels were in the early 1980s, only with a heavier texture to cover the inevitable irregularities in those Rek-O-Kut castings.

User avatar
jesusfwrl
Posts: 365
Joined: Thu May 16, 2013 2:24 pm
Location: Earth
Contact:

Re: RekOKut Imperial II restoration project

Post: # 45061Unread post jesusfwrl
Mon Dec 05, 2016 5:21 am

Grooveguy,
I was also wondering about the paint job on your Rek-O-Kut. Really beautiful indeed, I would say far better than the original. Is it powder coating that they did?
~~~ Precision Mechanical Engineering, Analog Disk Mastering ~~~
Agnew Analog Reference Instruments: http://www.agnewanalog.com

Post Reply