More G36 Questions!
Hi there,
Further dipping my toe in the dark and treacherous "moderizing a tube deck" lagoon, I'd like to convert one of my G36s to high speed (15/30 ips).
What I'm trying to figure out is the best way to achieve this. There is plenty of evidence that merely substituting a thicker (high-speed) capstan will cause an incorrect tape path. Is there any (practical) way to, say, reposition the existing tape guides to approximate a better path?
Alternately, I'm told that the "correct" way to move up to high-speed is to use a high-speed shaft AND high-speed motor, as it has a slightly different footprint to compensate for the thicker shaft.
Anyone know if this is all there is to it (I say "all" with a smirk, knowing how difficult it will be to find one)? In the G36 service manual, I see a number of parts that are designated for high-speed operation, including the capstan, capstan motor, and head support block. So I'm wondering if there's more to this than meets the eye.
Finally, what's the "definitive" story regarding 50 hZ vs. 60 hZ.? Is it impossible to usea 50 hZ motor in my (North American) deck? I'm told there may be a conversion possible, but I'm none the wiser.
Many many thanks, in advance!
Seth
Hi there,
Further dipping my toe in the dark and treacherous "moderizing a tube deck" lagoon, I'd like to convert one of my G36s to high speed (15/30 ips).
What I'm trying to figure out is the best way to achieve this. There is plenty of evidence that merely substituting a thicker (high-speed) capstan will cause an incorrect tape path. Is there any (practical) way to, say, reposition the existing tape guides to approximate a better path?
Alternately, I'm told that the "correct" way to move up to high-speed is to use a high-speed shaft AND high-speed motor, as it has a slightly different footprint to compensate for the thicker shaft.
Anyone know if this is all there is to it (I say "all" with a smirk, knowing how difficult it will be to find one)? In the G36 service manual, I see a number of parts that are designated for high-speed operation, including the capstan, capstan motor, and head support block. So I'm wondering if there's more to this than meets the eye.
Finally, what's the "definitive" story regarding 50 hZ vs. 60 hZ.? Is it impossible to usea 50 hZ motor in my (North American) deck? I'm told there may be a conversion possible, but I'm none the wiser.
Many many thanks, in advance!
Seth
No factory made G36 could do 30 IPS. Its a very tall order to get that transport running so fast.
It would be the playback head that you'd have to move. I suppose you could made a different adapter plate - the thing with the azimuth and tilt adjustment screws on it, and so accomodate a different head position.
The standard motor probably hasn't enough torque to run reliably at 15IPS. At 60Hz it might do better than the 50Hz one.
Electrically, to run it at a different line frequency you have to change the motor capacitors. It will then run , but at the wrong speed - so a different capstan is required - but the 50/60 difference is not so great as to require different physical positioning of heads and motors.
Probably easier to build an electronic sine synthesizer for the right frequency - just for the motor. The amplifiers won't care.
I did wonder once about trying to run the motor at 100Hz to see if it would do it. Did no experimenting though.
Thanks much for the info! Yes, I meant to say "15 ips"!
Come to think of it, someone else recommended modifying the head mounting plate. It so happens I have an extra, so maybe I'll try making a template from the current correct (but low-speed) one and marking how the tape is supposed to contact the playback head....
And thanks too for the motor info. Building a sine synthesizer sounds a bit beyond my skill level at the moment, but you never know. I have a line on a high-speed motor, but it's a) 50hZ and b) I don't definitively know what else would be required to fully convert to 15ips (though your answers give me a bit more certainty).
Thanks again,
Seth
Come to think of it, someone else recommended modifying the head mounting plate. It so happens I have an extra, so maybe I'll try making a template from the current correct (but low-speed) one and marking how the tape is supposed to contact the playback head....
And thanks too for the motor info. Building a sine synthesizer sounds a bit beyond my skill level at the moment, but you never know. I have a line on a high-speed motor, but it's a) 50hZ and b) I don't definitively know what else would be required to fully convert to 15ips (though your answers give me a bit more certainty).
Thanks again,
Seth
The stock capstan motor has insufficient torque for 15ips operation, in fact others have alleged that in cold weather it takes the deck 15 or 20 minutes to warm up enough to run properly at 7.5ips. Note that converting for 15ips operation also entails significant modifications to the record and playback EQ.
I like my G36 MKIII quite a lot, but decided long ago not to try to convert it for 15ips operation. Resources for the parts to do this conversion are listed in earlier posts, but I cannot attest to their continued availability.
Have you replaced the electrolytics and selenium rectifiers yet? Unless this is known to be a very low hour machine you should do this soon.
Hey, thanks much for the response...strangely enough, I was just reading your first post on this thread and wondering if your deck turned out to be 15ips or not. Seems like there are VERY few stock 60hZ / 15ips machines, if any. I have a hard-copy service manual and there are no parts listed for that configuration.
So sadly, without lucking into one of those rare machines myself, I'll probably have to abandon this project altogether, as I have no use for a machine that won't hit 15 ips.
It's not TOO hard to find 50hZ parts, but it seems like running just the capstan motor on a specialized power supply would be silly, at best.
There IS one more obvious option, which is making a sleeve to fit over the current capstan shaft. But I lack the mechanical / mathematical know-how to figure out how thick the sleeve would need to be (unless it's really just as simple as "Twice as thick as the unsleeved capstan shaft."
Many thanks,
Seth
So sadly, without lucking into one of those rare machines myself, I'll probably have to abandon this project altogether, as I have no use for a machine that won't hit 15 ips.
It's not TOO hard to find 50hZ parts, but it seems like running just the capstan motor on a specialized power supply would be silly, at best.
There IS one more obvious option, which is making a sleeve to fit over the current capstan shaft. But I lack the mechanical / mathematical know-how to figure out how thick the sleeve would need to be (unless it's really just as simple as "Twice as thick as the unsleeved capstan shaft."
Many thanks,
Seth
Last edited:
Maybe not so silly. Variation in line frequency exceeds the wow/flutter attainable by that transport. So if you wanted really accurate speed regulation, ordinary mains power isn't the best bet. Add-on quartz oscillators for synchronously driven turntables is quite a common tweak, no reason the G36 wouldn't benefit just as much.
Double the capstan diameter gives you double the speed, BUT the Revox capstans are machined to mind-bogglingly accurate tolerances - you'd need a specialised machine shop to be able to match the original spec.
If you're going from 50Hz to 60Hz there's an additional capstan diameter multiplier of 5/6 or 6/5 depending on the direction of modification.
A 15ips capstan on the standard motor will still overload the motor!
Last edited:
Thanks Steerpike,
I agree, seems like trying to juice the current motor would be a poor bet. Can you tell me a little more about modding the unit to run a high-speed capstan motor at 50hZ in a 60hZ machine?
Would the quartz oscillator you mentioned feed just the capstan motor? I know nothing about them, but it sounds promising.
In addition, I understand the phase shift cap needs to be changed, and I'm wondering if the mains power would be an issue (according to the service manual, there aren't separate motors for EU / US voltages).
Best,
Seth
I agree, seems like trying to juice the current motor would be a poor bet. Can you tell me a little more about modding the unit to run a high-speed capstan motor at 50hZ in a 60hZ machine?
Would the quartz oscillator you mentioned feed just the capstan motor? I know nothing about them, but it sounds promising.
In addition, I understand the phase shift cap needs to be changed, and I'm wondering if the mains power would be an issue (according to the service manual, there aren't separate motors for EU / US voltages).
Best,
Seth
As I said, I've thought about pushing a 7.5IPS motor up to 100Hz, since the increased frequency will give it extra torque, so it might cope. In your case, that would be a 120Hz modifiation. It is extra wear on the motor bearing though. I don't have any clue yet whether it will work, or if the magnetic properties of the rotor are incompatible with frequency doubling.
If you had a 50Hz motor/capstan, it will run too fast in the USA, so you'd have to lathe down the shaft, not add a sleeve. Needs very acurate engineering tools to do it right.
If you cannot get a 60Hz/15ips capstan motor, rather than metalworking, my choice would be to build a quartz reference 50Hz sinewave osc, and run it through a (chip) power amp into a step-up transformer, as the tuntable people do. The motor is low power - about 6W or so I seem to recall - so its not heavy current circuitry.
Building a direct coupled 220V (i.e., no step-up transformer) sinewave oscillator is much more difficult, so the chip-amp + transformer method is probably best to start.
All three motors need capacitor changes if you change from 50Hz to 60Hz; though if you run the capstan from its own synthesized 50Hz supply, the reel motors are still powerd by 60Hz mains and won't need any cap swapping. If you designed the oscillator suitably (a quadrature oscillator) you could omit the capacitor for the capstan motor altogether.
The 'normal' motors were 6/12 pole. There was a brief production run of high speed machines that used 4/8 pole motors and a correspondingly odd capstan shaft. These were an interim measure until a high torque 6/12 pole motor was produced. Just something to be aware of if you are searching for spare motors.
If you had a 50Hz motor/capstan, it will run too fast in the USA, so you'd have to lathe down the shaft, not add a sleeve. Needs very acurate engineering tools to do it right.
If you cannot get a 60Hz/15ips capstan motor, rather than metalworking, my choice would be to build a quartz reference 50Hz sinewave osc, and run it through a (chip) power amp into a step-up transformer, as the tuntable people do. The motor is low power - about 6W or so I seem to recall - so its not heavy current circuitry.
Building a direct coupled 220V (i.e., no step-up transformer) sinewave oscillator is much more difficult, so the chip-amp + transformer method is probably best to start.
All three motors need capacitor changes if you change from 50Hz to 60Hz; though if you run the capstan from its own synthesized 50Hz supply, the reel motors are still powerd by 60Hz mains and won't need any cap swapping. If you designed the oscillator suitably (a quadrature oscillator) you could omit the capacitor for the capstan motor altogether.
The 'normal' motors were 6/12 pole. There was a brief production run of high speed machines that used 4/8 pole motors and a correspondingly odd capstan shaft. These were an interim measure until a high torque 6/12 pole motor was produced. Just something to be aware of if you are searching for spare motors.
I may be missing something but surely you would expect much less torque at 100Hz due to the much reduced current. I wouldn`t be suprised if a high inertia motor like the Papst would never even run up to speed unless you increased the supply volts way beyond the safe limits.
Papst capstans are notorously hum prone and I would expect even more of a problem at 100Hz especially as hum bucking between the capstan, spool motors and mains transformer would be affected.
The value of the phase shift cap is critical at the lower speed to minimise flutter, reducing the supply slightly may also help.
On a different note, why do you want to run at 30ips anyway ?. Most professional studios standardised on 15ips when half decent tape became available in the late 1950`s.
Barry
Papst capstans are notorously hum prone and I would expect even more of a problem at 100Hz especially as hum bucking between the capstan, spool motors and mains transformer would be affected.
The value of the phase shift cap is critical at the lower speed to minimise flutter, reducing the supply slightly may also help.
On a different note, why do you want to run at 30ips anyway ?. Most professional studios standardised on 15ips when half decent tape became available in the late 1950`s.
Barry
Hmm, still trying to puzzle this one out:
-I currently have a 60hZ, low-speed machine.
-I could install a high-speed motor and capstan, but only 50hZ units are available (if at all).
-If I did, I would need to supply JUST the capstan motor with 50hZ voltage. that means, probably, building a quartz synthesizer, as Steerpike suggested. I've not yet encountered one, so I don't know how complex / how large / how expensive they are.
Is this insane? I'm beginning to suspect it may be.
-I currently have a 60hZ, low-speed machine.
-I could install a high-speed motor and capstan, but only 50hZ units are available (if at all).
-If I did, I would need to supply JUST the capstan motor with 50hZ voltage. that means, probably, building a quartz synthesizer, as Steerpike suggested. I've not yet encountered one, so I don't know how complex / how large / how expensive they are.
Is this insane? I'm beginning to suspect it may be.
Look for an Otari MX-5050 MKII or later and build yourself some good outboard tube electronics. The transports on these decks work like a dream.. Otari is still making decks new and parts are fairly readily available for the older decks.Ha! Yeah I think you said it Kevin. I actually have a couple of Otaris that I'm fond of, both are early-model 5050s, a 1/4" 2-track and a 1/2" 8-track. They are generally very solid pieces of equipment. I think I was just reeled in (seriously, no pun intended) by the sexiness of tubes.
Speaking of which, I know that some company (Bottlehead?) makes a tube amp replacement for the Otari, but it's darned expensive. I have built tube amps before, I wonder how hard it would be to fabricate them for the 2-track....
Oh no, better not get started down that path!
Many thanks,
Seth
Speaking of which, I know that some company (Bottlehead?) makes a tube amp replacement for the Otari, but it's darned expensive. I have built tube amps before, I wonder how hard it would be to fabricate them for the 2-track....
Oh no, better not get started down that path!
Many thanks,
Seth
Yes, thats what you'd need. The turntable people do it with things like this
Needle Doctor 1-800-229-0644, Pro-Ject Speed Box SE II
or the Linn Valhalla power supply.
Not that I imply that one will work with the G36, but the principal is identical. It would run ONLY the capstan motor. $600 is exceptionally expensive for what is inside it, but I suppose they feel they need to justify their time spent designing it. The fabrication cost is one 25W mono chip amp + power supply, plus an additional transformer. The 50Hz reference is then either your computer, an old MP3 player, or a quartz crystal plus sundry ICs (about $10 worth).
If you wanted to boldly go where no ReVox tweaker has gone before, you could put an A77 or B77 capstan motor + drive board into a G36. It's a far torquier motor. Everything is possible to those will it!
Jumping topic somewhat, does anyone have any information on the output transformer for the G36? The schematic says 6W into 5 Ohms, though the transformer seems a little overbuilt for these ratings. I was under the impression that it was servicing, essentially, a pair of EL84s, making it suitable for a Vox AC15-type circuit. But the ECL twin-package tubes are a little tamer? Any insights would be appreciated!
Hi Slor,
The 6BM8 differs from the EL84 in fundamental respects including various plate ratings so I would not consider them as being similar.
I believe electrically this is actually a pretty good transformer, the design I think intended to minimize coupling into the heads and surrounding circuitry. Might be good as a pair in a small 6BM8 PP stereo amp..
The 6BM8 differs from the EL84 in fundamental respects including various plate ratings so I would not consider them as being similar.
I believe electrically this is actually a pretty good transformer, the design I think intended to minimize coupling into the heads and surrounding circuitry. Might be good as a pair in a small 6BM8 PP stereo amp..
Actually it is something to seriously consider, I've heard the 5050 with bottlehead electronics.
Building a tube based tape replay amp is no more difficult than building a comparably complex phono stage.
My ReVox G36 is going to a new home very shortly.
In theory this seems the case, just using the NAB or IEC time constants, but unlike with riaa phono stages, how does one know what head compensation has been built in to the original replay amp?
I plan to build one for my Otari MX-55, and sell my G36 which is only 7 1/2 ips. Though the Otari has a wonderful smooth controllable transport, the G36 actually stacks the tape on fast rewind much better - like the Otari at 15ips (which the Otari doesn't at all).
It would be simple if one could input a rta sweep and pick it back up off the replay head, delayed slightly. Probably wouldn't work, but might be worth a try. Is there a programme that plots a sweep signal by itself? ie one might just digitalise the playback and look at it in isolation. as a sweep.