Is it possible to build a high-quality reel-to-reel tape recorder yourself? A beginner’s experience.

kris8888 wrote:

I think there might one day be a trend for watching ‘old’ films on VHS tapes (e.g. original tapes from former video rental shops). Contrary to what one might think, quite a few of these tapes have survived.
Given that there’s also a trend for analogue, primitive cameras (so-called ‘Lomo cameras’) or Instax-type cameras, anything is possible.
In any case, I no longer throw away any working CRT TV, video recorder or cassette player that I still have and for which there’s room in the garage.

I tried giving away original VHS tapes of children’s cartoons for free (my grandchildren don’t watch them any more) – nearly two hundred of them – but there was zero interest; I ended up taking them to the scrapyard (they were reluctant to take them). I’ve got old TVs, but I’m only keeping those from before 1965; the rest goes to the scrapyard; there’s zero interest, and if anyone does turn up, they suggest I send them by courier (I can’t be bothered), and the profit from that is at most enough for a four-pack; a PCL86 transformer is worth more than that. I’ve got a mobile electronics scrap yard right by my fence every Saturday.

And I’ve got two Amethysts to get up and running. I’ll definitely do one of them. So do move the date – it’s worth it.
Best regards

It’s not worth it....I’ve had all types of Ametyst (the easiest TVs to get working); I managed to get three of them up and running (a former Zurt employee, I’ve got all the parts for Polish valve TVs, especially those manufactured after 1968, including picture tubes), but nobody needs that rubbish these days. Try getting one working and selling it; then you’ll learn the reality of the market. I don’t want my sons to one day hire a skip and, whilst chucking out the rubbish, curse their father for cluttering up the workshop like that. Best regards.

cirrostrato wrote:

Nobody needs that rubbish these days

You can make a good mirror for a telescope out of a cathode-ray tube.

kaz69 wrote:

If you want to record a large amount of data digitally, wouldn’t it be better to use a video recorder?

As for digital recording, you could come up with something along the lines of DDC tape recorders, i.e. recording/reading using a stationary head to capture several slow data streams, which are then combined into a single main stream.
Since obtaining or constructing a precise, stationary multi-track head is impractical in an amateur setting, one could make do with two or three standard stereo heads from reel-to-reel tape recorders, arranged in the path of movement one after the other, with an appropriate vertical offset. The recordings from these heads must then be multiplexed and demultiplexed programmatically into a single common data stream.
Highly efficient codecs are now available which, even at bitrates of 96 kb/s or lower, provide quite satisfactory quality.

@andreyatakum As I’ve had quite a few tape recorders pass through my hands – starting with the East German Tonko – a rather nifty ‘tape deck attachment’ – as the name suggests – which was essentially a fully-fledged tape deck, albeit without a power amplifier or speaker, through a Soviet-era device similar to a Bruel & Kjaer measurement (reel-to-reel), a couple of Ampex units and ending with a Technics 1500 – and if I may suggest something to you, when you’re planning to build a reel-to-reel from scratch, stick to just three motors, as this will simplify the mechanics to a minimum. ,

Krzysztof Kamienski wrote:

Just three engines, as that will keep the mechanics to a minimum. ,

Exactly! That’s the route I chose a while back. Unfortunately, I didn’t finish the project – moving house, university, family.

kulmar wrote:

If we’re going to have a bit of fun, we might as well make the most of modern technology – the recording on the tape should be digital.

As far as I’m concerned, it’s complete nonsense! And how will this differ from a record player, sound-wise? Just because it’ll be a huge block with spinning reels? You’d think people would be doing this for the analogue sound!

kulmar wrote:

kris8888 wrote:

But it raises further questions. I suppose you realise how wide a bandwidth such a tape recorder needs to carry in order to be able to record and play back an audio signal in digital form?

Yes, I do. But I also see no reason to solve old problems in the ‘old’ way. For example, Kudelski seems to have recorded a rotation synchronisation signal on the tape. What’s to stop us from ‘embedding’ such a signal in the tape? And, for example, noise reduction systems – don’t we have the capability today to record the signal on tape with a constant dynamic range, whilst also recording information about the signal’s dynamic range on the tape, and then using that during playback to reproduce the dynamic range? Do we really want to simply copy what already existed without taking into account the capabilities of modern electronics?

I’ve thrown in a few other ideas – analogue ones. And you, as I understand it, want to recreate the ZK120 (without the T).

rosomak19 wrote:

kulmar wrote:

If we’re going to have some fun, we might as well make the most of modern technology – the recording should be digital.

As far as I’m concerned, it’s complete nonsense! And how will this differ from a CD player in terms of sound? Just because it’ll be a huge block with spinning reels? You’d think people would be doing this for the analogue sound!

But there are DAT recorders, aren’t there? So that’s digital recording on tape.

kris8888 wrote:

Perhaps CRT televisions?

If we’re going all out, then mechanical image reproduction based on a Nipkow disc.

Jogesh wrote:

But there are DAT recorders, aren’t there? So that’s digital recording on tape.

I know, but I listen to tapes for that analogue feel; and if I want digital, I’ve got a computer and CDs.

vodiczka wrote:

kris8888 wrote:

How about CRT televisions?

If we’re going all out, then mechanical image reproduction based on the Nipkow disc.

I tried that as a child too. I managed to get the ‘snow’ effect. There’s a DIY project available showing how to build such a television at home.

kulmar wrote:

If we’re going to have a bit of fun, we might as well make the most of modern technology – the tape recording should be digital.

Philips has already made such attempts. However, DAT very quickly proved to be a flop; in theory, digital technology offers the potential for extremely high quality (for magnetic tape recording) quality, but the problem lies precisely with the tape itself – whilst in the case of analogue recording, momentary drops are not a problem (as at most you’ll hear a crackle or a minimal drop in quality), with digital recording it wasn’t quite so straightforward. No error correction schemes could remedy the problems arising from the nature of the tape itself. Whilst the CD is still doing quite well, DAT was essentially forgotten within a few years of its launch.

Jacekj wrote:

Hello, I’d go for valve amplifiers.

What for? I suppose just to end up with higher electricity bills. A valve won’t improve the sound quality of a tape recorder. Modern silicon electronics already have such high specifications that, in this particular case, they’re ‘too good’ for magnetic recording anyway.

kulmar wrote:

Ultimately, it’s the medium – that is, the tape – that determines these parameters.

Yes and no. Whilst it is true that the tape has a significant impact on recording quality, noise level, bandwidth, etc., the influence of the mechanical aspects should not be underestimated; the precision of tape guidance, contact with the head faces, and irregularities in tape movement...

Despite its relatively simple operating principle, a tape recorder combines several extremely important fields – electronics, mechanics and … technology (the latter referring to tape manufacturing). You can have the best tape recorder in the world, but if the tape is poor, you will never achieve good sound quality. You can have the best tape, but if the electronics are faulty, you will not be able to achieve a good recording, let alone a good playback from the tape (even if recorded on a reference-quality tape recorder). Conversely, a minor fault with the pressure roller can compromise the quality to such an extent that a planned evening of music turns into a nightmare.One could list all these factors one by one and their impact on the sound, but there are so many that it would take up so much space that no one would have the patience to read through them all. And unfortunately, all these factors are crucial to achieving the ultimate goal: a sufficiently high-quality recording or playback. The best tape recorders cost a fortune (still do!) not because they’re big, heavy and look good, but because they have to meet all these requirements – and that, unfortunately, comes at a price.
And Nagra achieved its fame precisely because Mr Kudelski had the knowledge and the patience to see to every tiny detail and bring them to perfection.

andreyatakum wrote:

The Chinese are already offering tape-transport mechanisms.

Indeed, but which ones? The two or three simplest and cheapest mechanical models, which were mass-produced right up until the last ‘dachshund’ (radio cassette player) rolled off the production line. A chassis made from sheet metal as thin as a tin can, plastic flywheels (at best with a sheet metal disc attached), a permanent magnet instead of an erase head.... Even with the greatest determination and the best of intentions, the best you could make of this would be a poor-quality ‘Walkman’.

andreyatakum wrote:

There are more of these things gathering dust in readers’ cupboards and cellars than there are cassettes or records.

That’s right; unfortunately, cassette tape ages not only in terms of sound quality but also physically. I myself have at least about 600 cassettes (mostly branded ones)
and I know what I’m talking about – not to mention the findings from laboratory tests that were carried out back in the day. In theory, the tape retains its properties for a few to a dozen or so years at most; after that time, even when stored in its original packaging (sealed packs), it degrades, which affects its sensitivity, noise levels and overall quality. And to pre-empt the point that the Chinese are still producing cassettes – yes, they are, but they’re producing knock-offs. The recording quality on such tape, compared to a branded one (even one that’s been sitting around and has already deteriorated), is at best 20 per cent. The inherent noise is comparable to the recorded signal; the tape can start to delaminate even on the very first playback, and the tape itself can stretch.
Unfortunately, as I wrote, in a tape recorder it is important to maintain the best possible quality of each of its components – including the tape.

The tape itself isn’t the problem. With a wide tape and a fast feed rate, it’s perfectly possible to achieve good sound quality; for example, Hi-Fi video recorders easily cover the 20–20,000 Hz range.

Given the specifications of cassette recorders and their recording speeds, it’s clear that results can vary. Incidentally, I wonder what it would sound like if the tape speed were doubled.

398216 Usunięty wrote:

Yes, but which ones? The two or three simplest and cheapest mechanical models,

Unfortunately, that’s true. With a plastic flywheel. In my youth, I dealt with proper steel mechanisms and heavy flywheels on Soviet-made cassette players. But I wasn’t even satisfied with those; even Soviet manufacturers preferred to use imported mechanisms. These Chinese ones are a bit of an oddity.
However, the fact that they’re being produced shows there’s a demand for them. That’s what I wanted to say.

398216 Usunięty wrote:

However, DAT very quickly turned out to be a flop; in theory, digital technology offers the potential for extremely high quality (by the standards of magnetic tape recording) quality, but the problem lies precisely with the tape – whilst in the case of analogue recording, momentary drops aren’t a problem (as at most you’ll hear a crackle or a minimal drop in quality), with digital recording it wasn’t quite so pleasant

In my opinion, DAT recorders were mainly undone by the price of the cassettes themselves, which were quite expensive as they had to be made of chrome at the very least, and of the highest quality. Added to this was the fact that DAT recorders weren’t backwards compatible with standard compact cassettes, whilst there was also a lack of originally recorded DAT cassettes on the market. Furthermore, CDs eventually became widely available and affordable, and cheap CD burners appeared on the market, meaning that DAT essentially ceased to have a raison d’être.
I owned a Sony DAT recorder for a while and didn’t notice any drop-outs during recording or playback, even on cassettes that weren’t exactly new. I suppose the error correction and the quality of the cassettes themselves were good enough to cope with this. The problem, however, was the rather complicated mechanics of the recorder itself; I had issues with it. Mainly, it was down to the – unfortunately – plastic gears, which would crack over time.

And the latest floppy disks – Verbatim, Srim, etc. – the no-name ones were better. The former could only be written to once and there were problems reading them, particularly with track ‘zero’. In my opinion, the Chinese weren’t putting enough metal in them.
Best regards

James596 wrote:

Hi-Fi video recorders easily cover the 20–20,000 Hz range.

Hi-Fi video recorders record audio using heads on the video drum, which ensures the tape runs at a speed sufficient to ‘fit’ all the information. In the case of ‘normal’ tape (1/4 inch) and the recording system familiar from reel-to-reel tape recorders, it is a completely different matter. Even if a sufficiently high tape speed were achieved, it would be at the expense of recording time – a ‘studio’ reel could hold at most a few minutes of digital recording. DAT employed a trick involving the recording of a digital signal onto tracks ‘drawn’ by the video head drum. In this way, a cassette the size of a standard audio cassette could hold far more footage than would be expected from the tape’s specifications and its transport speed alone. Nevertheless, whilst everything looked great in theory, it soon became apparent that in practice things weren’t quite so rosy. All it took was a speck of dust on the tape, a hair or some other debris, and the whole elaborate plan went to the dog’s. I had an alternative to DAT, namely ‘DCC’ (the system’s originator) for three years, and suffice to say I used it very sporadically – preferring to rip the material onto an MD rather than risk playing it from the original DCC cassette and risking interruptions during the performances. Generally speaking, I believe that when it comes to digital recordings, SONY made the biggest mistake of its career by demanding licence fees from every potential MD manufacturer instead of (as Philips did with the CC) granting them the freedom to produce equipment under their own brands free of charge. As a result, only a few manufacturers decided to purchase a licence and produce MD players (and even then, only a small number of models). It’s a shame, because (following improvements to the ATRAC system) the quality of MD could have rivalled that of the CD, with the added bonus of all the extras MD offered – including the ability to record your own comments, titles or other information, and (a substitute, but still) your own ‘mastering’ of recordings.
So, if one might put it that way, DCC was a dead end, whilst the MD was the result of excessive ‘stinginess’.

andreyatakum wrote:

However, production figures show that there is demand.

There is a demand because a few smaller manufacturers (particularly Chinese ones) decided to meet consumer expectations and began producing radio cassette players or even ‘Walkmans’. Unfortunately, this does not alter the fact that their quality was in no way comparable to that of cassette recorders from their heyday. Customers bought them – partly out of nostalgia, partly because the recorded cassettes remained whilst the equipment had been scrapped, and in order to at least recall the days of their youth, they needed to get hold of something that would allow them to play back the recordings.

kris8888 wrote:

In my opinion, DAT recorders were mainly undone by the price of the tapes themselves, which were quite expensive because they had to be made of at least chrome and of the highest quality.

The price was indeed quite high, but the equipment itself (on paper) offered much more than a standard tape or cassette. All in all, it was meant to be an alternative to professional-grade equipment, and that was that. And the prices of blank cassettes (because at the time, equipment of that quality was something you could really only read about) were an incentive for those seeking high quality. It didn’t work out, because it couldn’t for the reasons described above. This included (as you yourself wrote) the complicated (because it was miniature compared to VHS) mechanics. That’s a shame.

398216 Usunięty wrote:

that their quality was in no way comparable to that of cassette recorders in their heyday.

I know; I can see that from the photos of the mechanisms.

398216 Usunięty wrote:

In this way, it was possible to fit far more recordings onto a cassette the size of a standard audio cassette than would be suggested by the tape’s specifications and its playback speed

Perhaps not ‘a lot more’ recordings, because as far as I recall, depending on the length of the tape in the cassette, a DAT cassette could hold between 60 and 90 minutes of music at normal sampling rate. So, much like a compact cassette.
The difference was simply that a much larger amount of digital data was stored on a DAT cassette than would have been possible with a standard compact cassette and a regular cassette player.

I suppose another reason the DAT format failed was that there were practically no portable devices like the Walkman. Given the mechanical components and the rotating head, it would have been difficult to make anything small. And I don’t think there were any in-car DAT players either. Discman players were much more readily available back then.

The MD digital format, on the other hand, was an interesting alternative, especially for portable devices. However, once again, due to audiophiles seeking high-end equipment for the home, the MD format was largely overlooked because of its lossy method of sound recording.

You’re right. I got a bit confused with that ‘much larger quantity’ bit… Sorry. However, I think (my memory can be a bit dodgy) I vaguely recall seeing a portable DAT player, one that could even record…but I wouldn’t swear to it, because I might be getting something mixed up again.
As far as I remember, when MD players first appeared, there wasn’t really that much talk about audiophiles yet. In any case, the general consensus was that the CD was the format offering the highest possible quality, and as the MD was similar in terms of recording (optical on a disc), nobody really had anything bad to say about it. Anyway (I remember this quite clearly), back when I was still working in a repair shop, I had a few customers with MD players, and from their conversations it seemed they were the sort of people for whom the price of a power cable was of paramount importance. A lossy format – yes, MD never made a secret of that, but whilst even I had reservations about the early ARTRAC systems, the later versions were actually very good (I repeat – in my opinion).
Perhaps, however, we’d better get back to tape recorders...

398216 Usunięty wrote:

Perhaps we’d better go back to tape recorders, though...

Actually, why bother? What can we improve? We’ve been there, done that.

kulmar wrote:

But why, actually?

To get back to the topic?

kris8888 wrote:

And I don’t think there were any car DAT players either

I think Blaupunkt made one, but it was only available for a very short time at the turn of the 80s and 90s.

This is how I’d see it –

I’d copy the separate write and read circuits, as well as the bias current generator, from the ZRK M7020, for example, I’d copy the BIAS control from the Koncert – initially using switchable BIAS, as in cassette recorders, via isostats or relays, and eventually introducing automatic correction

He didn’t go for classic mechanics as in the heyday of reel-to-reel tape recorders, but did it this way – the drives for the reels and pressure rollers were implemented directly using stepper motors / BLDC motors controlled by an ATmega. Here’s an interesting concept for stabilising the tape speed: we have a disc with a circumference of 4.75 cm fitted with a small permanent magnet. When the magnet passes through the Hall effect sensor, it generates a pulse.
And so: 1 pulse = 4.75 cm/s
A pulse half as long = 9.5 cm/s
A pulse a quarter as long = 19 cm/s
A pulse an eighth as long = 38 cm/s
The figure 4.75 cm/s is not chosen at random – using simple maths – by multiplying by 2, 4, 8 and 16, we can obtain 9.5, 19.5 and 38 cm/s (small studio version) and 76 cm/s (studio version)
The microcontroller reads the pulses from the Hall sensor and compares their durations – and controls the motors via the drivers accordingly so that the specified speed is achieved. This greatly simplifies the mechanics, as braking can also be implemented using an ATmega. In that case, we are limited to three motors, two solenoids for lifting the bridge, and a few integrated circuits

I wonder whether stepper motors from printers would be suitable?

How does a stepper motor work? Here’s the answer to your question:

CHOPIN66 wrote:

Would stepper motors from printers be suitable?

The tape feed MUST be smooth, without any ‘jerks’ or jolts. Even the slightest such ‘glitch’ affects the listening experience – what’s more, a high-quality tape recorder specifies, amongst other things, two parameters that define this – ‘WOW’ & ‘FLUTTER’, and these parameters must not exceed the permissible limits – ideally, they should be imperceptible. Furthermore, jerking of the tape during playback, recording or rewinding can (and most often does) cause the tape to stretch. Not only does stretched tape ruin the recording, but it also starts to flake – the magnetic layer applied to the tape itself begins to peel off. Since you’ve already used the ‘Koncert’ as a benchmark (incidentally, a rather poor choice), why didn’t you choose the tape recorder on which the ‘Koncert’s’ mechanism is based, namely the REVOX? Forgive me, but in my humble opinion, you’re a classic example of someone who wants to reinvent the wheel. Unfortunately, the development of the tape recorder is a rather lengthy process, with all manner of ‘setbacks’ along the way to perfection – or at least towards that ideal. During this process, there were revolutionary designs, and some that didn’t work out; there were also total flops – but as has always been the case, each subsequent model incorporated solutions that proved successful, whilst discarding those that caused problems. And there were, and still are, a great many such problems to be solved when it comes to the tape recorder as such; starting with the design of the head itself and extending to the method of achieving the smoothest possible drive. What’s more, even a copy of a good solution isn’t necessarily perfect – the Koncert, for example, was admittedly modelled on the mechanical solutions of the Revox A77, but not entirely – several discussions (not only on Elektroda) have been written about it, and unfortunately many of them are not particularly positive. Leaving aside the flaws in the mechanics themselves, the electronics unfortunately leave much to be desired (particularly the automatic control systems).
I hope I haven’t dampened your enthusiasm – I merely wanted to demonstrate that even when modelling a design on a ‘ready-made’ product, one must be able to identify those elements that have a decisive influence on the final result; and to be able to spot them, you unfortunately need to know far more than just general information about how something works – you need to know why it works the way it does, what you should pay particular attention to, and what can be omitted or altered. As my late grandfather used to say – the road to hell is paved with good intentions… explaining the meaning behind this proverb: “Good intentions are fine, but knowledge is the foundation.”

CHOPIN66 wrote:

Here’s an interesting concept for stabilising the belt’s feed speed – we have a disc with a circumference of 4.75 cm fitted with a small permanent magnet. When the magnet passes through the Hall effect sensor, it generates a pulse.
So: 1 pulse = 4.75 cm/s
A pulse half as long = 9.5 cm/s
A pulse a quarter as long = 19 cm/s
an 8-times shorter pulse = 38 cm/s
The figure of 4.75 cm/s is not chosen at random – using simple maths – by multiplying by 2, 4, 8 and 16, we can obtain 9.5; 19.5 and 38 cm/s (small studio) and 76 cm/s (studio)

You haven’t exactly discovered America here; this type of capstan motor speed stabilisation was used in some tape recorders, mainly reel-to-reel models. It is also used in direct-drive turntables. Sometimes it’s not a Hall effect sensor but a simple coil resembling a miniature read head, but the principle of operation is the same.

>>21926676
They used to be around. I’ve got a SONY DAT somewhere in my storeroom – not much bigger than a standard Walkman, but heavy and sturdy. It played and recorded, and was used as a reporter’s piece of kit. Whether it still works today – I don’t know; the last time I fiddled with it was about 20 years ago...

398216 Usunięty wrote:

The tape feed MUST be smooth, without any ‘jerks’ or jolts.

Yes. You’re absolutely right, and that’s exactly what I was going to write in my reply to my colleague. But powering the stepper motor from a sinusoidal oscillator will solve this problem. It could be the same oscillator I used to drive the turntable platters.



Not exactly the same one, but one operating on the same principles – using a Wien bridge amplifier. For example, high-power integrated circuits. You just need to recalculate the frequency depending on the characteristics of the stepper motor being used.

An Arduino controlling the power amplifier via PWM (pulse width modulation) could also work, but I’d avoid that. I like to build analogue receivers with digital interfaces (synthesisers, F and S-modulation meters), including on Arduino, and I’ve had to struggle with interference from these interfaces. So I wouldn’t recommend using digitally controlled tape-winding motors.
However, when it comes to the flywheel drive (especially with quartz stabilisation) and controlling the entire system, there’s no other option here. You just need to use shielding and ferrite cores.