Skip to main content

Calling all veteran engineers!

I've got an Akai GX-4000D 1/4" reel-to-reel machine I picked up from Ebay. It's a great unit with nice clean sound and in fantastic condition.. BUT..

I have a very strange problem whereby the playback speed seems to be fractionally faster than the record speed! The symptom is basically that when I bounce a track from my DAW down to tape and then back into the DAW, the track that I record back into the DAW is a little shorter (and therefore I'd assume higher pitch, although the difference is small enough to be inaudible). That means of course that it falls out of time with the rest of the track. The same goes if I record straight to tape and then "import" it into the DAW.

Has anyone ever experienced such a thing? I've checked the tape path and there's no extra heads in contact with the tape during recording or anything. I would assume that it would have to be exactly the same servo driving the reels for playback and recording so it's all a bit of a mystery to me. I'm not scared of wielding a screw driver and hammer on the thing if required but just thought I'd check here first. I'm a little bamboozled as to what could cause it but I don't have any previous experience with tape machines.

Thanks all.
Matt

Comments

JoeH Sun, 07/31/2005 - 23:21

If it's that small a difference (say, a second or two over the course of a three minute song), you may not be able to get it any better than what you've got. I believe you have a "consumer" machine, or "Pro-sumer" at best there.

It's an analog device, remember. And it's not something you can use an external sync source with, like today's more accurate Digital machines. You can't run it as a slave; it's got its own internal reference - be it an oscillator or AC line reference, and it will drift a bit throughout the course of a song.

If anything, your playback speed (motor) needs adjustment. But again, it may not be worth tweaking. It could be as good as you'll get it. If your capstan & tape path is in good shape, and if your take up and supply tension is set up properly, it may be as good as it gets within the spec of that machine.

If you DO decide to go after it and you're feeling brave, you'll need a frequency counter, an alignment tape, and the service manual for your machine.

The service manual (probably available somewhere online if you google the model #) will show you where the motor speed alignment trim pot is (either on a circuit board, or perhaps even on the motor itself; I don't remember on your model....) Some models have separate rec and pb speed, but it's probably just one trim pot for all.

The frequency counter will let you read the alignment tape's pitch accurately to make sure you're running at the right speed.

The Recorder/Reproducer Alignment tape will provide you with an accurate Playback Speed reference. (Make sure you're using a specifically created playback SPEED tape, not just alignment tones.)

But be aware, many of those trim pots aren't very reliable, and the smallest of tweaks can create a big jump or change in pitch. (Some folks modify the the stock pots with something with more turns and a higher resolution for precisely this issue.)

If you're just running onto tape (rec head) and back out (repro head) at the same time, then you'll probably not have a problem...just use the machine as an aux send loop, and get the bennies of tape warmth & distortion. You wont have tape speed issues here, you'll be a second or two delayed, (which you can slide in your DAW) but it'll be accurate overall if you're able to rec/play back simultaneously.

Otherwise, you'll have to accept the fact that most analog machines (unless they're truly "pro" machines running with an external controller/sync mechanism) will not run completely in sync for more than a few seconds or so...eventually the "Drift" happens, and after a few minutes, it's quite noticeable if you're A/B'ing it to a DAW track.

JoeH Mon, 08/01/2005 - 09:46

It's a consumer 1/4" reel to reel machine, from what I can see.

Here's a link to a similar one on sale on EBAY, and it's WITH a service manual. (Possibly worth as much as the machine now.) This one is 240/250 Volts AC - a UK Machine, FWIW.

http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&category=109002&item=5792664567&rd=1

Judging by the picture and description, it's from the late 70s' (I thought it was more recent than that, I was confusing it with one from the 80's, I guess...) All the more reason to suspect he wont get the speed lock/adjustment he's looking for.

From the product description on Ebay:

AKAI 4000DS : Reel to Reel Tape Recorder - Working Condition :: NR

History :

Akai 4000DS Mk2 machines where manufactured circa 1976-1978 and they recorded and played back giving excellent results (due to the 1 micron gap heads).

The Mk 2 models were the successor to the 4000DS (circa 1972-1976) which in its turn was a successor to the famous 4000D (circa 1970-1972) with several improvements. These machines proved to be reliable workhorses like their forebearers, this was due to the simple but rugged construction which gave a build quality beyond its competitors in the same price range.

The last "4000" model produced by Akai was the GX - 4000D. This machine incorporated glass/crystal ferrite recording heads that resulted in a better frequency response and boasted a lifetime guarantee!

Indeed, these machines from Akai are still noted for their reliability and endurance which has made them something of a legend, they are still going strong after 30 years plus.....how many modern machines will be able to boast that?

...Have tested the basic features such as play, fast forward, and rewind which seem to be working. I am not expert enough to know about the more technical features of the machine. The small screw-on metal cap that allows you to change speed is present. The VU meters both work and light up. There are a few minor marks and scratches on the wooden parts or the metal front (which could do with a good wipe).

Serial Number : S-90102-02084

Included with the tape recorder :

Two reels (one full, one empty),
One Original Spool Retainer,
Wall Plug,
Phono Lead,
Photocopy of 43 page Service Manual for 4000DS / 4000DS-Mk2.
A few other empty plastic tape reels
I believe the Original Technical Specification was :

Track Format: 4 track, 2 channel stereo/mono
Reel Size: up to 7’’ spools
Tape Speeds: 7.1/2 and 3.3/4 ips
Wow and Flutter: less than 0.15% RMS at 7 1/2, less than 0.2% RMS at 3 3/4
Signal-to-noise ratio: better than 50dB
Erase Ratio: better than 70dB
Bias Frequency: 100 kHz
Cross-Talk: better than 70dB (mono), better than 50 dB (stereo)
Heads: (3) 1 micron gap playback head, 1micron gap recording head, 1 erase head
Fast forward and rewind time: 150 seconds using a 1200 ft. tape
Motors: 1 4-pole induction motor
Power requirement: 220-240V, 50Hz
Aprox Dimensions: 40 x 30 x 15 cms

This item is very heavy - as a result, the P&P charges will total £17.50 within UK. Free Local pickup is available, from Ashwell, near Baldock, Hertfordshire, or Stevenage, Hertfordshire.

*No Reserve*

CoyoteTrax Mon, 08/01/2005 - 11:01

I have a couple of old open reel decks I use for their beautiful preamps and sometimes insert as line amplifiers. But as far as re-amping or tracking to tape on them and then bringing that track back into my Daw...I've always had the same problem you're describing.

RTR's rarely playback at the exact same speed unless they've really been professionally setup. Mine are so old there's not alot I can do about that so I haven't bothered but you may be able to have your Akai serviced to playback more accurately (mine are from the 50's and 60's).

In the end though, with a consumer deck like that (high speed = 7 1/2 ips) you may never be able to sync that machine well enough to do what you're trying to do in your DAW.

Hopefully Chance or someone else has some advice on how to make that happen.

In the mean-time, the Akai pre's should provide very interesting color to any of your mic's and as a line-amp it should provide some interesting color as well.

anonymous Tue, 08/02/2005 - 09:55

The only open reel units that I trust to be accurate from record to playback are Studer and Otari, but I do have the pres from other "prosumer" models. I sold a Sony TC-388-4 on eBay about 6 months ago with the line "works great for parts, missing preamps". Anyway, just wanting to add that if its the AKAI pres you like the sound of, just do like the others have said and use them and not the tape.

anonymous Tue, 08/02/2005 - 13:52

Thanks for all the info guys!! That's basically cleared everything up for me and saved me some hours of pain with a screwdriver..

I'm not a huge fan of these pres, I'm sure I'll be able to use their colour for something but the main thing I wanted was to get some tape saturation on some tracks and experiment with recording direct to tape.

I guess the way I can use it now is with some sort of pitch shift but then I lose quality so no go really.

Oh well. I'll start saving for the "real thing" and just use it as and where I can for bouncing down entire cuts that don't mind an extra couple of cents higher pitch!

Thanks again guys,
Matt

JoeH Wed, 08/03/2005 - 01:40

Matt, I do believe you can get this effect without the tape drift problem if you use the Akai as an effects loop in real time, but NOT as a playback device.

In other words: If you record signal onto tape in one pass, and play it back on a second pass, then, YES you will get the pitch/tempo drift you described.

BUT, if you set it up as essentially a tape delay running in real time (By taking signal from the tape/playback head WHILE IT'S RUNNING IN RECORD MODE), you wont get the pitch drift. (A little phasing/flanging perhaps, but nothing terribly bad.) You'll get a several millisecond time delay from the difference between the rec & Playback heads, but once it's all finished you can of course nudge it forward again on the timeline to match the rest of your tracks.

If you don't have simultateneous play/record functions with your computer sound card there is another method, albiet somewhat tedious.

First, burn the tracks you want to use onto a CDr, or make a DAT (or any other highly reliable non-drift playback method - preferably digital.)

Then, play the CDr (or DAT, etc.) into your Akai in record mode, with tape running.

Lastly, put your computer in record mode, taking the TAPE (NOT the source!) output of the Akai.

Doing it this way, you'll get a real-time playback from the CDr (Hopefully in much better sync) being printed on tape, but without the time/pitch changes in a stand-alone analog tape deck running without sync in PB mode.

I used to do this (even with a 3-head cassette deck, once, when a reel to reel wasn't available) to 'warm up' some cold, lifeless vocal tracks a client did with eary digital gear. We were too impatient to record it all once, and then play it back a second time, so we simply did the "Rec/repro" instant tape delay method.

Should work like a charm for you, if you set it up correctly.

kostas Mon, 04/30/2018 - 11:34

Matt_Trix, post: 160334 wrote: Thanks for all the info guys!! That's basically cleared everything up for me and saved me some hours of pain with a screwdriver..

I'm not a huge fan of these pres, I'm sure I'll be able to use their colour for something but the main thing I wanted was to get some tape saturation on some tracks and experiment with recording direct to tape.

I guess the way I can use it now is with some sort of pitch shift but then I lose quality so no go really.

Oh well. I'll start saving for the "real thing" and just use it as and where I can for bouncing down entire cuts that don't mind an extra couple of cents higher pitch!

Thanks again guys,
Matt

Hi Matt, i have an Akai 4000ds mkii and i have the same problem. Is your model with 3 heads the one next to the other? If it is you can record on the tape and back to your daw at once. This way not envovle the play speed, only the record speed. I hope that is helpfull

ronmac Mon, 04/30/2018 - 15:16

Kurt Foster, post: 456756, member: 7836 wrote: has anyone ever noticed that digital recorders don't cite a wow and flutter spec? does any one know why?

Because digital recorders don't have wow and flutter issues.

Digital recorders all lock to a very accurate clock, either internally or via an external word clock or smpte string, offering precise sampling during record and playback.

Not all clocks are created equally, but even a very poor clock today will "jitter" only at a rate of a few millionths or tens of millions of a second. Some manufacturers post jitter specs, but most don't as it is (arguably) not worth mentioning.

KurtFoster Mon, 04/30/2018 - 15:43

ronmac, post: 456757, member: 24337 wrote: Because digital recorders don't have wow and flutter issues.

Digital recorders all lock to a very accurate clock, either internally or via an external word clock or smpte string, offering precise sampling during record and playback.

Not all clocks are created equally, but even a very poor clock today will "jitter" only at a rate of a few millionths or tens of millions of a second. Some manufacturers post jitter specs, but most don't as it is (arguably) not worth mentioning.

..... and that is why you can never wild sync any RTR to a digital recorder without a sync tone or clocking the analog machine from the digital clock.

wow and flutter causes analog recorders (all of them) to drift when they are played against a digital source. we used to be able to get away with about 10 to 15 seconds when "wild syncing" a chorus back ground vocal or some other repetitive part. we would mix a short section of a part to a 2 track and then drop it back in on choruses or heads. it would usually take a few attempts to get one right.

paulears Tue, 05/01/2018 - 00:14

HANG ON! While it's true that reel to reels may well not run at exactly 7 ½IPS, or 15IPS - this doesn't;t matter a jot when you have just one machine. For your length difference, it must have changes speed BETWEEN recording and replay, which is extremely unlikely. If it slowed down or speeded up it would be detectable normally. Speed changes were a real pain when you have multiple machines and just one of them is out, even by a tiny bit. If your machine ran at 8IPS, you'd never know, until deck number 2 came along.

So you need to check the stability, and that is easy to do. generate a tone in your DAW - something like 1K would work pretty well. Record this to the Reel to Reel, then replay it. First thing is to check the tone is stable and has no drift up or down in tone frequency you can detect by ear. Then record that back into the DAW. If playing back the two produces a pretty obvious phasing sound, then your reel to reel has a speed control issue - the Akai from memory has a servo speed control that uses feedback from the motor to ensure stability. This could be broken or faulty. If your ear can't hear any problems with the two tones - one DAW, one from the recorder, then the errors must be smaller. If your DAW has a phase meter display - handy for checking stereo content, use this. You should see on the tones a steady straight line in most meters. If it starts to rotate, it indicates the tone from the recorder is changing slightly. Again - pointing to a servo error of some kind where it's slowing up or down.

4000 series recorders in the mid 70s, were damn good machine, and very reliable. It could be something as silly as a drive belt slipping, or the capstan motor motor and servos, but servo errors get 'averaged' by the flywheel to a degree.

If you can prove the speed stability is the cause, I'd look for perished drive belts to the capstan roller first - if this seems ok, then it's going to be servo based (that's assuming my memory for the machine is still sound) - it's been a long time since I had one of these machines (I was about 17!)

Boswell Tue, 05/01/2018 - 02:33

Using a tape machine as an insert device in a channel of your audio interface completely eliminates any effects due to steady-state tape speed errors. In addition, wow will not be noticable when using a high enough tape speed, as the record-replay delay is less than 100 milliseconds. The only concern might be flutter, and, in my experience, this is almost unmeasurable when using pro machines on a high speed setting.

Whenever I've used this method in a recording session, I've spent 10 mins at the start of the session recording sinewaves from a generator though the tape insert at approximately 100, 1K and 10K Hz. It's important that these frequencies are not exact multiples of one another. At mixdown, I use the sine waves to get an accurate measure of the time slippage of the tape-inserted track(s), and then move the affected tracks forward in time to compensate (to the nearest sample). I always recorded the track that temporarily had the generator on it as both direct (no tape insert) and via tape, and in that way I could check the time alignment using the phase diagram as Paul mentioned. While it was usually possible to make out a small amount of unsteadiness in the phase plot due to the mechanical process involved in the tape insert path, it was usually so small that it required a trained eye to detect.

dvdhawk Thu, 05/03/2018 - 07:56

paulears, post: 456762, member: 47782 wrote: HANG ON! While it's true that reel to reels may well not run at exactly 7 ½IPS, or 15IPS - this doesn't;t matter a jot when you have just one machine. For your length difference, it must have changes speed BETWEEN recording and replay, which is extremely unlikely. If it slowed down or speeded up it would be detectable normally. Speed changes were a real pain when you have multiple machines and just one of them is out, even by a tiny bit. If your machine ran at 8IPS, you'd never know, until deck number 2 came along.

So you need to check the stability, and that is easy to do. generate a tone in your DAW - something like 1K would work pretty well. Record this to the Reel to Reel, then replay it. First thing is to check the tone is stable and has no drift up or down in tone frequency you can detect by ear. Then record that back into the DAW. If playing back the two produces a pretty obvious phasing sound, then your reel to reel has a speed control issue - the Akai from memory has a servo speed control that uses feedback from the motor to ensure stability. This could be broken or faulty. If your ear can't hear any problems with the two tones - one DAW, one from the recorder, then the errors must be smaller. If your DAW has a phase meter display - handy for checking stereo content, use this. You should see on the tones a steady straight line in most meters. If it starts to rotate, it indicates the tone from the recorder is changing slightly. Again - pointing to a servo error of some kind where it's slowing up or down.

4000 series recorders in the mid 70s, were damn good machine, and very reliable. It could be something as silly as a drive belt slipping, or the capstan motor motor and servos, but servo errors get 'averaged' by the flywheel to a degree.

If you can prove the speed stability is the cause, I'd look for perished drive belts to the capstan roller first - if this seems ok, then it's going to be servo based (that's assuming my memory for the machine is still sound) - it's been a long time since I had one of these machines (I was about 17!)

I don’t disagree with any of that, but I think it overlooks another important variable.

Sure, consistent capstan speed is a big factor and as pointed out is generally very stable on any decent machine. The mechanism of any tape-based machine has to include one other thing that’s a lot harder to regulate - slippage. Due to the ever changing circumference of the tape, on both the supply-side reel and the take-up reel there has to be a certain amount of slippage engineered into them.

On newly loaded tape the empty take-up goes like a bat out of hell, while the supply barely turns. They’re both going the same speed in the middle of the reel and have traded roles by the end. The take-up mechanism can’t logarithmicly adjust its RPMs throughout, and it can’t pull the tape any faster than the pinch roller will allow, so it has to slip.

This constant tug of war on both ends would account for more wow and flutter than inconsistent capstan speed - given the rubber drive parts are good.

paulears Thu, 05/03/2018 - 09:47

If the capstan and pinch wheel are in good order then they don't slip. All the better class machines, and that's pretty well anything that takes NAB spools has decent tape tension compensation and the meters we used to use would measure it, and allow it to be perfectly fine. If you put your finger on the take up spool and let the tape spool out, then let go, when tension is taken up there's a bang and a lurch and you hear that badly as it rips the tape through the capstan/pinch wheel gap - but that's a transient - in normal use speed stability is rarely caused by the supply and take up spools going mad, and any even, consistent speed increase is from the capstan running at the wrong speed. Any slip in that area is huge and horrible and wouldn't cause a decent recording to not match up when brought back in. That can only happen if replay speed is different to record speed. The spools don't slip, they have a motor and a braking mechanism - either direct drive, or an out and back hub brake, with a solenoid to do a full stop brake, and an adjuster with perhaps felt pads to create a gentle back pressure on the supply spool, and either again direct drive or a clutch on the take up spool.

I remember very well having to adjust these things, and getting it just right was pretty hard - but it doesn't impact on tape speed unless it's windy wrong and then the horrible noises make that very obvious. Revox, Studer, ferrograph and the ASP modded ones all took a great deal of workshop time. When I qualified back in the late 70s, our speciality in the workshop was Ferrographs - I came in at Series 7, and they went through the last few models while I was there. Revox and the others for me came a bit later, and I was never as good as on the Ferrographs. The super 7 was my favourite reel to reel ever! Wish I'd never sold the one I boughtt.

paulears Wed, 05/09/2018 - 13:42

As many tape decks can drop in and out of record by a button - there is no difference in tape tension when recording or playing. Friction has no impact on tape speed on a machine that is set up and operating correctly - as we've said. tape speed is controlled by the capstan and pinch wheel. If this is set so that the spool motors can pull it through, they do so at a crazy rate - maybe 30ips plus, not close to the normal speed. The capstan must be changing speed if the replay speed is not the same as the record speed - why is this? Presumably it's simply faulty and the speed control is not working properly - usually the servo monitors the speed it's turning and if it is going to fast slows it down and if too slow speeds it up - so quickly the speed is to all intents and purposes constant. The problem we are talking about here is NOT friction, or slippage, it's loss of capstan stability.

DogsoverLava Wed, 05/09/2018 - 17:04

paulears, post: 456905, member: 47782 wrote: As many tape decks can drop in and out of record by a button - there is no difference in tape tension when recording or playing. Friction has no impact on tape speed on a machine that is set up and operating correctly - as we've said. tape speed is controlled by the capstan and pinch wheel. If this is set so that the spool motors can pull it through, they do so at a crazy rate - maybe 30ips plus, not close to the normal speed. The capstan must be changing speed if the replay speed is not the same as the record speed - why is this? Presumably it's simply faulty and the speed control is not working properly - usually the servo monitors the speed it's turning and if it is going to fast slows it down and if too slow speeds it up - so quickly the speed is to all intents and purposes constant. The problem we are talking about here is NOT friction, or slippage, it's loss of capstan stability.

but wouldn't we see the variability occur in both modes (record and play?) And why not friction if record mode (which also "plays") brings the record head in contact with the tape causing additional tension/friction -- could this not account for the variation happening in record mode only?

KurtFoster Wed, 05/09/2018 - 18:05

the record head is in contact with tape regardless if it's in playback or record. many recorders have only one head for both playback and record.

wow and flutter causes very slight random changes in speed. even a record turntable will exhibit the same behavior. anything that is not clocked will show very slight variations. [[url=http://[/URL]="

"]early Warner Bros. film sound locked a record / disk with audio to picture by driving the turntable by gear off the projector.[/]="

"]early Warner Bros. film sound locked a record / disk with audio to picture by driving the turntable by gear off the projector.[/]

DogsoverLava Wed, 05/09/2018 - 20:28

Kurt Foster, post: 456908, member: 7836 wrote: the record head is in contact with tape regardless if it's in playback or record. many recorders have only one head for both playback and record.

wow and flutter causes very slight random changes in speed. even a record turntable will exhibit the same behavior. anything that is not clocked will show very slight variations. [[url=http://[/URL]="

"]early Warner Bros. film sound locked a record / disk with audio to picture by driving the turntable by gear off the projector.[/]="

"]early Warner Bros. film sound locked a record / disk with audio to picture by driving the turntable by gear off the projector.[/]

I did not know that (about the record head). Thanks for expanding my understanding there. Makes me wonder about residual magnetization.... is that a thing?

paulears Thu, 05/10/2018 - 00:04

If you buy a tape deck on Ebay with a fault that means it records and replays at 10" per second - if this is your only tape deck, and you don't swap tapes with anyone else, you will NEVER notice it.

Residual magnetisation is a real, and very important thing. Years ago, every serious tape user, amateur and professional had a degausser. You regularly cleaned the heads, guides and capstan with isopropyl alcohol, then demagnetised the heads. The magnetisation actually starts to erase a tape when played, gradually reducing it's level every play, and the magnetisation also prevents full depth recordings, and the sound suffers, especially at the top end where the depth of tape saturation is lower. You also need a calibration tape that you use to make sure your tape head gap is at exactly 90 degrees to the tape edge. Again, if this drifts off through vibration, or rough handling, it needs adjusting. The starting position at a recording studio meant you spent ages each day cleaning and aligning all the recorders so they could swap tapes happily from machine to machine. Cleaning and form filling was a really big part of the job, but an amazingly important one.

You can buy degaussers on amazon and Ebay, and they're not expensive - but beware, as incorrect application can increase the magnetisation if you don't do it properly.

One really good thing you can do is to record some test tapes. They will be subject to any problems your own machine has, and of course will be your own standard - replay on somebody else machine not aligned the same will not works so good.

Record some tones. pick a low middle and high. Adjust the record level to hit the 0VU marking on the meter for each one. On replay you can check the purity of the sound. Wow and flutter will be very obvious. Level changes will also show up. Ideally replay of each tone will show you the frequency response is constant. There will be differences. Keeping a record will show you drift - usually the HF is coming down. Cleaning and degaussing indications. White noise is also quite useful for adjusting the azimuth - the brightest and cleanest top end is when the head gap is perpendicular to the tape. If you never share tapes with another machine. Having your own unique alignment is not necessarily a bad thing.

These test tapes will also let you keep an ear on other issues. Some designs have a clean tape path. The supply spool has a constant torque, and the tape is held tight against the heads by the steady back tension and the capstan/pinchwheel just drives it through. These machines tend to have a gentle curve to the two or three heads top assist the pressure. Other designs may have little levers with tiny felt pads that push the tape into contact with the heads. The steady tones often allow you to hear a kind of fluttery sound as these pressure pads skip and hop over the surface when their springs are weak.

My essential tool kit for a reel to reel was a small screwdriver to adjust the head azimuth, a degausser, bottle of alcohol, cotton buds and a lint free cloth. A cocktail stick is also handy for removing bits of built up oxide from the guides. Sadly, if I was buying a reel to reel machine now, I'd have to add a jewellers eye glass to be able to get my eyes close in - my glasses would be no good whatsoever!