Discrete Vs Combi Cassette Deck Heads

[knobtwiddler]

Hi,

I didn't want to derail the thread on tape types, so thought I'd ask here:

Mitajohn wrote:

Yes Revox B215 is a superb cassette deck as it's professional brother Studer A721, equipped with transformer balanced Ins/Outs and parallel RC. But both they have a little flaw: While they are 3 headed decks, the R/P head block is constructed in such a way not permitting azimuth adjustment of the record head! You had to live with the manufacturer's set-up. If you were lucky your head block would be close to "a few" degrees of azimuth error between the record and reproduction heads.

So...if you want the best from the B215, you need 2 of them: one aligned for record, and the other aligned for play?

I have a collection of precious tapes and am also thinking of making a few new ones for my own amusement. I'm teetering around getting a high-ish end deck (at a daft time - while cassette is the height of hipsterdom).

The latter Dolby S-equipped Sonys tick most of the boxes (aside from my partner not being enamoured with their looks - she is more partial to a Revox or Technics and says a Sony can live in the workshop). A discrete deck would be great, but many of the nice ones I've looked at do not handle metal, and some of my tapes are.

I was planning to get a 215 until I read John's comment. Which has gotten me thinking that all 3-head machines that have combi heads must have the same issue... I could probably find a Nak (discrete), but I'm put off them as I've read that servicing can be tricky. The 215 ticks all the boxes, other than the azimuth issue John describes.

One attraction to the 215 is that if I ever needed a new head, I have a NOS Amorphous Canon Combi Head (looks similar to ones I see on later Sony decks).

Thanks in advance!

[Restoration73]

I would find that head compromise would be unacceptable. The main purpose of the separate playback head is for live monitoring. Some manufacturers e.g. Aiwa, Denon,
used a separate play head which just fitted in the cassette body opening just left of
the capstan. However when the recording was finished the main rec/pb head was
used for playback as in most 2 head decks.

[paulsherwin]

You only need metal tape support if you are going to record on metal cassettes. The playback characteristics are the same as pseudochrome.

[TIMTAPE]

Originally Posted by knobtwiddler

Hi,

I didn't want to derail the thread on tape types, so thought I'd ask here:

Mitajohn wrote:

So...if you want the best from the B215, you need 2 of them: one aligned for record, and the other aligned for play?...

For the best reproduction I guess yes, but what a price to pay for it!

In their sales brochures Nakamichi pointed out the azimuth alignment problem in standard "sandwich heads" and rightly claimed that their decks were the only ones with the separate adjustment for the record head. But as we know, some of those companies took notice and the improved heads pictured earlier were made. I wonder if later in the series, Revox/Studer fitted such heads as well.

Tim.

[knobtwiddler]

Considering that discrete heads have so many theoretical advantages, why were most decks (bar-Nak) using combi 3-head (separate erase head, PB + Rec in same housing/ mount) from the late 80s onwards? The very last Sony decks not only had Dolby S, but all kinds of audiophile extravagance - I.e. dual mono layout of electronics (the PCB CAD dept had obviously never seen cassette head crosstalk data...) and separate transformer taps for different functions. If they lined the thing with copper cladding, why not go the whole hog with discrete heads?

Thank you for that info, Paul. It's decades since I had a deck with manual bias buttons (current pair tell me what they think the tape is). I thought Metal demanded a specific replay EQ, but you are saying FE/CR position will do the trick (assuming deck has that switch).

I went through more tapes over the weekend. I have Denon and Yama decks at present. In truth, I doubt a more elaborate deck will sound better as I am limited by the record quality from 30 odd yrs back (some of the old non-HX Marantz recordings sounded better than I thought they would). As Paul highlights, the question is whether i want to relive my youth and make metal recordings... Sound wise, the current pair of decks probably suffices well. The 215 I had for a while seemed to sound better on tapes made by the Denon... but that could be my memory (or an EQ disparity, giving the Studer more top end). The Denon has HX (Studers didn't, right?).

If I find a new, old deck, it'll be more as an objet d'art than anything. Nice meters and a satisfying clunk when I want to enjoy old tapes. If I want to record with decent quality, I suspect a later Sony or B215 will be the way to go (Nak is too much of a headache - and I suspect my spare amorphous Canon head will fit a 215 or late Sony if that ever were an issue)

[TIMTAPE]

Originally Posted by knobtwiddler

Considering that discrete heads have so many theoretical advantages, why were most decks (bar-Nak) using combi 3-head (separate erase head, PB + Rec in same housing/ mount) from the late 80s onwards?...

Azimuth misalignment between tapes and machines generally was a problem not only with compact cassettes but other slow speed tape formats. Open reel tapes at 1 7/8 ips and slower were fussy re audio azimuth alignment, especially half track. Professional full track 1/4" recordings usually didnt go below 7.5ips partly because of azimuth unreliability. VHS and Beta cassette audio was also very azimuth sensitive. So in that sense the small misalignment you sometimes found within the combi head unit wasnt unusual or huge, but it meant a 2 head machine playing back its own recordings could outperform a 3 head in that department.

[Analogue man]

I would have hoped that in the case of the Revox/Studer machine that company's build quality would have assured that the two elements of the dual head were azimuth matched i.e. the two gaps in the core were parallel both with each other and with their neighbours.

[TIMTAPE]

Originally Posted by Analogue man

I would have hoped that in the case of the Revox/Studer machine that company's build quality would have assured that the two elements of the dual head were azimuth matched i.e. the two gaps in the core were parallel both with each other and with their neighbours.

I would hope so too but there are manufacturing tolerances regardless of the status of a company. We're talking maybe fractions of a degree. Re stereo tape heads the service manual often mentions that the left and right gaps may be at slightly different azimuth angles, as well as not perfectly time aligned even if azimuth aligned, due to manufacturing tolerances, and the best we can do is strike a compromise or mean between the two gaps - and that's within the one head.

Interestingly in the Tascam 122 MkII and Mk III pro cassette deck service manual the instructions for head azimuth alignment only refer to the single azimuth screw which adjusts the azimuth of both record and play heads together. But included is a small diagram which identifies the separate record head azimuth screw. It seems they saw the record head azimuth adjuster as mainly for correcting azimuth misalignment due to manufacturing tolerances.

[paulsherwin]

One thing to bear in mind is that the compact cassette system was designed as an inherently two head system from the very beginning, and all three head arrangements are arguably compromises or even bodges. While there are theoretical advantages to seperate rec and pb heads, the benefits are small apart from easier auto calibration and off tape monitoring.

[jamesperrett]

Originally Posted by knobtwiddler

I thought Metal demanded a specific replay EQ, but you are saying FE/CR position will do the trick (assuming deck has that switch).

It is actually the 70uS chrome replay setting that you need for metal tape playback. I seem to remember that FeCr tapes also used that equalisation setting but decks that have a FeCr setting are fairly rare nowadays.

[mitajohn]

Originally Posted by Analogue man

I would have hoped that in the case of the Revox/Studer machine that company's build quality would have assured that the two elements of the dual head were azimuth matched i.e. the two gaps in the core were parallel both with each other and with their neighbours.

Years ago I asked Studer about this azimuth matching problem. They warrantied an azimuth error of less than 90 degrees at 10kHz.

For your info: A Tapeheads.net forum member suggested a replacement head Sony RPA 230-3602 amorphous head with separate R/P azimuth adjustment and mechanically adaptable on B215.

[Ted Kendall]

Originally Posted by TIMTAPE

Professional full track 1/4" recordings usually didnt go below 7.5ips partly because of azimuth unreliability.

In the 1960s, BBC Bush House converted 150 Leevers-Rich tape machines to 7.5/3.75 ips to obtain greater tape economy, the thinking being that nobody would notice the quality difference on shortwave transmissions. Unfortunately, the chunky deckplate wasn't quite chunky enough - if you set the azimuth at 3.75, raised the deckplate and lowered it again, it went out of adjustment. And this was before the studio managers kicked up about the impossibility of editing. Said machines spent the rest of their working lives at 7.5.

[TIMTAPE]

Originally Posted by Ted Kendall

In the 1960s, BBC Bush House converted 150 Leevers-Rich tape machines to 7.5/3.75 ips to obtain greater tape economy, the thinking being that nobody would notice the quality difference on shortwave transmissions. Unfortunately, the chunky deckplate wasn't quite chunky enough - if you set the azimuth at 3.75, raised the deckplate and lowered it again, it went out of adjustment. And this was before the studio managers kicked up about the impossibility of editing. Said machines spent the rest of their working lives at 7.5.

That's depressing. Stefan Kudelski designed his Nagra III and onwards with 3.75/7.5/15 ips speeds but perhaps anticipating the azimuth issue I think he limited the upper record frequencies on 3.75ips to less than what that speed was strictly capable of so that azimuth misalignment wouldnt be as audible.

Something which caught my eye recently was the IASA TC06 standards manual for analogue videotape digitisation which included slow speed consumer formats like Beta and VHS. Reams on getting the picture as true to the data on the tape as possible, but not a word on adjusting the linear audio/ctrl head to the tape's recorded azimuth pattern. Yet in the the same organisation's standards publication for analog audio digitisation, the importance of adjusting azimuth to the pattern on each audio tape is emphasised, as it should. I'm not sure what is going on there. Hopefully it will be rectified in a revised edition.

[knobtwiddler]

That Sony head appears to be the proverbial rocking horse manure from what I can make out. The last one appeared on Ebay in 2016.

I wonder what the members will make of the attached combi-head? It's branded Canon and is supposed to be of the amorphous type. It's unused. My feeling is that it's likely of an ilk that was used on mid-price Denon / Pioneer type decks, possibly in the latter stages of cassette machine production, when 3-head decks became affordable. Stats for it are as below:

REC Amorphous

R=3.0 R=48,1
L=3.0 R=47.6

PB Amorphous

R=110mH R=324
L=110mH R=324,2

I was thinking that were I to get a B215, this head could make a decent replacement *should* the need arise (I'm aware that many more problems are attributed to heads than is actually the case).

Cheers,

[TIMTAPE]

Originally Posted by paulsherwin

One thing to bear in mind is that the compact cassette system was designed as an inherently two head system from the very beginning, and all three head arrangements are arguably compromises or even bodges. While there are theoretical advantages to seperate rec and pb heads, the benefits are small apart from easier auto calibration and off tape monitoring.

Yes although on the other side, there were a few happy coincidences in the original Philips compact cassette design which left open the possibility of a three head setup.

Philips made the cassette 'turnoverable' mean the other set of holes for capstan and pinch roller on side 2 could be exploited for a dual capstan design.

The two "small windows" could accomodate a tiny erase head.

With some tricky design, completely separate record and play heads could just be squeezed in to the space normally occupied by a single record/play head.

Then Nakamichi worked out there was enough clearance to place a small bracket on the play head to just lift the cassette's pressure pad off the record and play heads for better tape to head contact, less scrape flutter and less head wear, and disabling the pad was only possible with a dual capstan to now tension the tape against the heads.

Nakamichi's decks were often difficult to set up and repair but I feel their going the whole way with the pressure pad lifter and physically separate and adjustable record and play heads were important features compared to other manufacturers' designs. My ideal 3 head deck would probably be a non Nak (for more reliability and easier servicing) but with all the important Nak features especially in the heads area.

[stevehertz]

I've often wondered why someone didn't make a deck where a loop of tape was drawn from the cassette video machine style, allowing all manner of physical layouts and improvements.

[knobtwiddler]

A poor man's Pressure Pad Lifter would be to spend a couple of minutes removing the pad in the cassette -) You'd just have to remember not to play the cassette in a machine equipped without dual capstans.

The very last Sony models rivalled the Naks for W+F, but had HX Pro as well as Dolby S. One imagines spares are easier to find than with a Nak as they were some of the last high-end decks from any OEM. I have seen people knock the Sonys for sounding too clinical and a bit like CD. Seems to me that Sony did their job too well. In a format that doesn't exceed 60dB SNR without electronic aids, I'd take 'clinical' as a compliment.

[stevehertz]

Originally Posted by knobtwiddler

A poor man's Pressure Pad Lifter would be to spend a couple of minutes removing the pad in the cassette -) You'd just have to remember not to play the cassette in a machine equipped without dual capstans.

The very last Sony models rivalled the Naks for W+F, but had HX Pro as well as Dolby S. One imagines spares are easier to find than with a Nak as they were some of the last high-end decks from any OEM. I have seen people knock the Sonys for sounding too clinical and a bit like CD. Seems to me that Sony did their job too well. In a format that doesn't exceed 60dB SNR without electronic aids, I'd take 'clinical' as a compliment.

I would class a 'clinical' description as being borderline audiophool territory.

[knobtwiddler]

If the B215 has the issue where azimuth is inconsistent between the Pb / Rec sections of its combi head, then doesn't that mean all decks that used combi heads have the same issue?

From what I can make out, all of the major cassette deck OEMs (with the exception of Nak) transitioned over to combi heads in the mid 80s. Sony made their last discrete decks around 86 (if I am right?). This begs the question: was the move away from discrete solely economic, or were there perceived benefits to the combi heads at the time?

Obviously, there is less hardware to go in the deck, and less tweaking on the production line. But the B215 uses a combi head, and there are no signs that Revox tried to save beans on it (witness the 4-motor mechanism). If Sony were transitioning away from discretes, then maybe that drove Revox to follow suit?

90 degrees strikes me as a lot of phase shift, if relative between L/R.

>the benefits are small apart from easier auto calibration and off tape monitoring

The only benefit I'm aware of for discrete 3 heads is being able to get the azimuth spot on for each one, as opposed to the Revox 'compromise'. A combi head will still let you monitor audio and set bias live. I'm assuming there are other benefits to discretes, due to the distance between them?

[TIMTAPE]

Originally Posted by knobtwiddler

If the B215 has the issue where azimuth is inconsistent between the Pb / Rec sections of its combi head, then doesn't that mean all decks that used combi heads have the same issue?

Not necessarily inconsistent, just unadjustable. We already established that some combination heads put out by Teac and Sony did provide separate adjustment of the record head azimuth. It can be useful after wear on a head or other issues in the tape path alter the azimuth alignment. In the studio Tascam 122's I'd occasionally check the azimuth between record and playback and it was rarely completely stable from tape to tape. Azimuth error in stereo was one thing with cassettes but in mono (or summed stereo) it was far worse. Try it and see for yourself.

Originally Posted by knobtwiddler

From what I can make out, all of the major cassette deck OEMs (with the exception of Nak) transitioned over to combi heads in the mid 80s. Sony made their last discrete decks around 86 (if I am right?). This begs the question: was the move away from discrete solely economic, or were there perceived benefits to the combi heads at the time?

I understood that only Nak adopted discrete record and erase heads. I dont understand how the non Nak OEM's moved away from something they had apparently never adopted.

Originally Posted by knobtwiddler

Obviously, there is less hardware to go in the deck, and less tweaking on the production line.

Yes and they may have been able to make height and tilt adjustments visually without having to install and set it up with a tape in a machine. Many reversing cassette decks used a combination erase/record and play head. The two heads were glued together. The same with many VCR's which allowed for audio dubbing. But of course the azimuth setting of an erase head is not that critical.

Originally Posted by knobtwiddler

But the B215 uses a combi head, and there are no signs that Revox tried to save beans on it (witness the 4-motor mechanism). If Sony were transitioning away from discretes, then maybe that drove Revox to follow suit?

Again AFAIK they didnt transition away from discrete. They never adopted discrete 3 head. Up to that point Revox had never made a cassette deck and probably had never made cassette deck sized heads. For their first effort I can see it would have made sense to leave the cassette head manufacture side to a company like Sony which had much experience in that department. Making high quality heads is one thing but making them that small as well would have been even harder especially in the company's first entry into cassettes and with everything else about the new deck very high quality. Why not farm out the heads side of things to a specialist?

Originally Posted by knobtwiddler

...The only benefit I'm aware of for discrete 3 heads is being able to get the azimuth spot on for each one, as opposed to the Revox 'compromise'. A combi head will still let you monitor audio and set bias live. I'm assuming there are other benefits to discretes, due to the distance between them?

Again, with wear in the tape path, and even using a different cassette, azimuth can change. Even within one side of a cassette, the azimuth often changes, partly because the reel tensions change. Azimuth was a huge problem with cassettes and I'm not sure even Nak ever completely sorted it but they probably did better than anyone else.

Nak also claimed that the distance between the record and play heads reduced crosstalk from the powerful record head to the sensitive playback head.

The high treble is obviously very sensitive to azimuth error, but also very sensitive to bias level. On top of that, Dolby tracking is very sensitive to treble error, so with Dolby the errors easily snowballed.