UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Tape (Audio), Cassette, Wire and Magnetic Disc Recorders and Players

Notices

Vintage Tape (Audio), Cassette, Wire and Magnetic Disc Recorders and Players Open-reel tape recorders, cassette recorders, 8-track players etc.

Closed Thread
 
Thread Tools
Old 12th Oct 2017, 9:26 pm   #1
REDGUUZ
Triode
 
Join Date: May 2017
Location: Assen, Netherlands.
Posts: 13
Default REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Dear fellow Tape Enthusiasts,

I'm overhauling a REVOX A700 Tapedeck (1975), which I own since 1999.

It has by now failed three times:
2006: Blown FRAKO Capacitor ad rectifying bridge in +12V power supply;
2013: Started smoking heavily during play back but worked All Right next day and no more smoke (no action from my part: insufficient time); It turned out later (2017) to be an exploded 0.1uF RIFA cap across the mains.
2017: most recent failure: Inability to rewind: (another) blown RIFA Cap and burnt safety resistor on the Motor Relay Board and associated Fuse F2: Rewind motor AC supply).

I decided now to do things properly and fully overhaul the deck, which is still occasionally used.
BTW. I was fully unaware that the above failures are well known; (I had not done any Internet investigation at all).

By now, I have already replaced all 10 RIFA Noise suppressor caps in the deck (these tend to shortcircuit and smoke) and all the Gold FRAKO ElCO's (these tend to dry out and shortcircuit). Most of them had only 1/3 of their nominal value.
Tantalum caps will still be replaced by WIMA MKS2 (film) 50V film capacitors ( < 10 uF) or new tantalum caps with a higher voltage 35V (except the large (> 10 uF) coupling caps in the audio circuits, for which I will use Nichicon FG electrolytic caps).

By now, I also have done a detailed study of the diagrams of this deck (and its sisters: STUDER A67 and B67 Mk1) and have studied various Internet Fora (www.Tapeheads.Net, www.audiokarma.org, www.studerundrevox.de), so I am now more aware of the various well known problem areas of these decks.
However, I have noticed that the following issues have been barely addressed:

I noticed that in the A67 and A700 STUDER was very skimpy with the specification of the operating voltage of the FRAKO caps and certain Bridge rectifiers in the Power supplies and motor driver boards (partly rectified in newer REVOX A700 models and the STUDER B67Mk1). I have therefore installed higher voltage (and current) replacements , just to be sure.

But I have still several questions;

1) In the Tapecontrol board (Attachment 1) there are 2 Motor Drive IC's, (TCA561), each followed by a transistor (BC108B) which each drive a Power Darlington (MJ411) and a reel motor (wind/rewind).
These BC108B transistors sometimes fail (destroying the (by now) unobtainable TCA561 driver IC); Source : "TINMAN": STUDER expert on the Tapehead.net forum.
Note: in the STUDER B67 a heavier Duty BC140-10 (Higer Vco 60V, higher I 800mA) has been applied.

My question : Could I not better replace the BC108 (Vceo=20V), Q17 transistor by a repacement with a higher Vceo (say 50-60V) ?
Ideally I would also specify a higher current and power rating: what transistor would you recommend? The BC140-10 used in the B67 has a lower Hfe, but maybe this not that important in this application.

(EDIT: on the DIYAUDIO.com site I got the recommendation] of the BC337 (NPN). . If I look at the site of my local parts shop the BC 639 (NPN) could be used as well, in my opinion. Would you agree??

2) Same question for Q5 (BC108) in the Capstanboard (attachment 2)

3) I noticed that the STUDER B67Mk1 uses BFR18 (Vceo = 60V) transistors in exactly the same locations of the Tape control IC (SC10429) whereas in the REVOX A700 BC107 's (Vce = 45V) are used.
(compare Attachmen3 with Attachment 4).
Is the BFR18 overkill?? In my opinion a BC107 would be adequate for the voltage (24V) employed.

4) I noticed that in the A700 a BC109C (Vceo=20V ), Q1 is used for the optical shut off circuitry, whereas the STUDER B67 uses a BC107 (Vceo = 45V), which, again, would be OK for the voltage (24V) employed.
The BC109C seems to be underdesigned.
Can I simply exchange this BC109C by a BC337 as well?

5) The A700 Deck employs a lot of BC107 transistors, so I do not understand why at certain places BC108B 's were used (were they much cheaper at the time (1975??).

All advice would be welcome!

Thanks for the help!

Martin
Attached Thumbnails
Click image for larger version

Name:	Drivertransistor Q17 BC108B.jpg
Views:	1081
Size:	60.6 KB
ID:	150652   Click image for larger version

Name:	Capstanboard  onderbemeten Q5 BC108B.jpg
Views:	814
Size:	56.6 KB
ID:	150653   Click image for larger version

Name:	Tapecontrol IC REVOX A700.jpg
Views:	425
Size:	70.3 KB
ID:	150654   Click image for larger version

Name:	Tapecontrol IC STUDER B67.jpg
Views:	707
Size:	53.9 KB
ID:	150655  
REDGUUZ is offline  
Old 12th Oct 2017, 10:17 pm   #2
paulsherwin
Moderator
 
paulsherwin's Avatar
 
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,787
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

The BC107 is exactly the same as the BC108 except for the higher working voltage so can be substituted everywhere. The BC109 is a low noise BC108, and again a BC107 can be substituted where the noise performance isn't critical.

The modern version of the BC107 is the BC547. This is a cheap and easily available transistor.

The BC337 is a different transistor but will probably also be OK. It has a higher collector current capacity than a BC547.

The BC639 has even higher voltage and current ratings than the BC337. It will probably be OK as well, but I have no personal experience with it. The BC639 has a bce leadout, unlike the cbe leadout of the other types mentioned here.
paulsherwin is online now  
Old 13th Oct 2017, 9:37 am   #3
ricard
Octode
 
ricard's Avatar
 
Join Date: Mar 2006
Location: Lund, Sweden
Posts: 1,631
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Quote:
Originally Posted by REDGUUZ View Post
My question : Could I not better replace the BC108 (Vceo=20V), Q17 transistor by a repacement with a higher Vceo (say 50-60V) ?
Ideally I would also specify a higher current and power rating: what transistor would you recommend? The BC140-10 used in the B67 has a lower Hfe, but maybe this not that important in this application.
I would say it depends on whether the transistor fails because the voltage gets too high or because it gets too hot (i.e. the power rating is insufficient).

The BC140 is a bit of a halfway house between a small signal and power transistor. It's got a high Vceo but it's still in a standard (albeit slightly larger) metal can so it can't handle that much continuous power. I wonder if a BD139 (1.5A / 80V) wold have an even better margin here. Like other power transistors, it does have a low Hfe though.
ricard is offline  
Old 13th Oct 2017, 10:51 am   #4
paulsherwin
Moderator
 
paulsherwin's Avatar
 
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,787
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

The point is that changing a BC108 to a BC107/BC547 has absolutely no electronic implications for the circuit. If you change to a BC337, BC639 or any other different type there may be implications, though this seems unlikely having quickly looked at the circuits.

I should state that I have no direct experience of working with these control circuits and am certainly not a Revox/Studer expert.
paulsherwin is online now  
Old 13th Oct 2017, 11:41 am   #5
REDGUUZ
Triode
 
Join Date: May 2017
Location: Assen, Netherlands.
Posts: 13
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Paul & Ricard,

Thank you for your replies/suggestions,

Since the BC108B (s) at the output of the TCA561 (s) only fail (s) occasionally, I think it is due to a too high Voltage (Vce=20V is too meagre!) given the application.
It does not have to dissipate that much power because after all it is followed by a MJ411 or MJ413 power darlington, which control the reeling motors.

The low Hfe of the BC140 (or the even lower Hfe for the BD139) might be an issue (maybe not) and can only be determined by Trial and Error.

I think the upgrade to a BC337-140 (Vceo = 45 V, Icont = 500 mA, P= 625 mw, Hfe = 400) versus BC108B (Vceo = 20V Icont = 100 mA, P = 300 mW, Hfe = 450) should be sufficient. This was also suggested on the Dutch Forum [https://www.circuitsonline.net/forum/view/138222/1/a700.
Maybe the BC639, Vco =100v, I cont = 1000mA, P=830mW, but Hfe=150 is not so appropriate. It might be overkill for Volt, Current and Power ratings but does not work due to a too low Hfe (maybe it is not an issue in this application).

(I would simply like to swap these parts now that I am "recapping" the machine, especially the Tape Control and Capstan Boards.
Maybe it is not necessary after all (so far all tantalum caps which I already replaced were within specs, but if one shortcircuits it could possibly destroy an now unobtainable IC (e.g. SC10429, TCA561, TDA1000), hence my desire to do it now properly instead of "Do not fix it, if it ain't broken" attitude sofar.

If anybody on the Forum has further suggestions for upgrading the REVOX A700, I would very much welcome input!!

Best regards,

Martin
Attached Files
File Type: pdf BC 109.pdf (53.5 KB, 131 views)
File Type: pdf BC337.pdf (56.7 KB, 77 views)
File Type: pdf BC639.pdf (51.4 KB, 91 views)
REDGUUZ is offline  
Old 13th Oct 2017, 11:54 am   #6
paulsherwin
Moderator
 
paulsherwin's Avatar
 
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,787
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

I doubt if the slightly lower hfe of the BC639 would be significant in this application. It isn't too far from a BC108B (200-450). However, it is always safer to make a substitution where this has already been done without problems. If others have switched to a BC337 then I would choose that - it gives a significant increase in voltage and current handling. The BC639 is probably overkill.
paulsherwin is online now  
Old 21st Oct 2017, 4:22 am   #7
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

"If anybody on the Forum has further suggestions for upgrading the REVOX A700, I would very much welcome input!!"


Coming a little late to this thread. I've owned an A700 for about 15 years and have had no transistor failure problems. The only failure was one tantalum capacitor which went short circuit and caused problems with the transport control. I think I also replaced the Rifa caps as a matter of course.

The only mod I made was to replace the (to my ears) very noisy TBA 931 repro amp with a lower noise substitute to bring it on par with the performance of similar less expensive machines like the A77, B77 etc which used discrete low noise transistors. Such a discrete circuit would probably be impossible with the A700 as there is very little room for it inside the special shielded metal box.

Tim
TIMTAPE is offline  
Old 24th Oct 2017, 12:13 am   #8
Michael Maurice
Moderator
 
Michael Maurice's Avatar
 
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,219
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

I think the reason they used the TBA931 is because on switching off, the +12V supply falls quicker than the -12V. That could induce a DC current which could magnetise the heads.
__________________
Forum Moderator

http://www.michaelmauricerepairs.co.uk/
Michael Maurice is offline  
Old 24th Oct 2017, 8:41 am   #9
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Yes but to me it still doesnt explain why in the A700 they used the TBA 931 which is demonstrably noisier than the cheaper A77/B77 model designs, and when for the later B77/PR99's they stayed with the standard discrete circuit. As mentioned the only reason I can think of is that there wasnt enough room in the A700's metal box to fit a discrete circuit.

Does anyone know why the 931 would be less vulnerable to magnetising the repro head due to the different rail voltage decays? Is there a modern 8 pin IC that would have this characteristic as well, while also having lower noise?
TIMTAPE is offline  
Old 24th Oct 2017, 9:59 am   #10
Ted Kendall
Dekatron
 
Join Date: Jul 2008
Location: Kington, Herefordshire, UK.
Posts: 3,657
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Dear me, this is an old story, so let me tell it again.

The A700 used cutting-edge technology to produce a machine with advanced characteristics, like a constant linear tension tape transport, at a price the domestic market would support. At launch, the machine cost £649 in the UK, as opposed to £350-odd for the A77, depending on the exact version. The Studer A80 was nearer £10,000.

To produce this complexity of machine for this sort of price necessitated the use of ICs wherever practicable, and for the most part this policy was completely successful. True, the odd chip goes down and replacement, over forty years later, can be problematical, but it was still a remarkable achievement, only six years after the A77, to produce a machine as far in advance of that excellent design - and let us not forget that the A77 was universally hailed as something different in kind, not degree.

I coveted an A700 from when they came out, and it took me several years to acquire one at a price I could almost afford - swiftly followed by two "well loved" examples. Perforce, then, I learned how to fix them - Bauch service cost a mint, excellent though it was, and to be fair, the service manager at the time, Mick Elliot, was generous with advice.

There were a few bugs in the A700 design - signal breakthrough in the record switching, popcorn noise, oddities with the autostop - but these were all addressed. The replay noise wasn't as low as it could be, but was well below tape noise nonetheless. The head works straight into the TBA 931, which provides most of the gain in the rep amp, so its noise level determines the overall s/n ratio of the playback electronics. To get adequate noise performance, it was best to use the -3 version of the chip, which was selected for low noise, and in practice this was satisfactory, unless you were using lots of top lift to retrieve a tape seriously lacking in treble - it was still well below tape noise in normal use.

Now, the TBA 931 went out of production, and a service replacement was required for chips which failed. Revox produced a plug-in replacement which was a 5532 on an adaptor board to fit the original 14-pin pinout. This worked perfectly well - until I tried it in the rep amp. It took me a while to work out that, although the rep noise was now lower, the rep head was getting magnetised every time the machine was switched off. This was news to Bauch, and indeed to Regensdorf, and some weeks later a service bulletin was issued to the effect that the substitute should not be used in this position. (There was somewhere in the A720 preceiver where it wasn't recommended either, if memory serves). The reason lies in the high input bias current of the 5532, combined with the rail architecture of the machine. The audio supplies are 12v pos and neg, but there is also a +5v rail for the logic, which is derived from the +12v rail. This asymmetrical load causes the +12v rail to decay faster than the -12v rail, and a large current to flow through the rep head.

So that's the problem, what about solutions? Well, a low noise chip with low input bias current - something MOS with the correct optimum noise impedance, maybe. Alternatively, a modification to the power rails to ensure a more symmetrical decay to the supply to this chip, or a discrete buffer as used on the Studer A810.

And what have I done? As yet, I freely confess, nothing - I have an A810 for noise-critical use, and one day...but for all that, I wouldn't be without 'em.
Ted Kendall is offline  
Old 24th Oct 2017, 10:36 am   #11
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Thanks for that extra detail Ted.

I must check which version is my A700, hence the TDA931 version fitted. From memory it's a fairly early vintage A700 and perhaps that explains why the hiss from the 931 is much worse than I've come to expect from say an A77/B77/PR99. Some time ago I posted a thread on the mod here, including some photos. I've even forgotten the type of IC I fitted but remember it was an 8 pin type.

Tim
TIMTAPE is offline  
Old 24th Oct 2017, 11:18 am   #12
Ted Kendall
Dekatron
 
Join Date: Jul 2008
Location: Kington, Herefordshire, UK.
Posts: 3,657
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

LM833, if memory serves - which has a listed input bias current higher than the 5532, if the sheets I consulted are correct. Does it work in your machine?
Ted Kendall is offline  
Old 24th Oct 2017, 12:27 pm   #13
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Yes with from memory about 6db lower average noise. However the machine is rarely used and I havent got around to testing the degree of possible magnetisation. In lieu of that, whenever I do use it, I demag the tape path before switch on, which I would do in any case.

Tim
TIMTAPE is offline  
Old 24th Oct 2017, 11:45 pm   #14
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Quote:
Originally Posted by Ted Kendall View Post
... To get adequate noise performance, it was best to use the -3 version of the chip, which was selected for low noise, and in practice this was satisfactory, unless you were using lots of top lift to retrieve a tape seriously lacking in treble - it was still well below tape noise in normal use...
I would have thought that, so long as the repro amp doesnt add significant extra noise to the baseline of bias noise, the lack of treble in the recording and so the need for top lift would be irrelevent here. Or did you mean a case where there is unavoidable spacing loss?
TIMTAPE is offline  
Old 25th Oct 2017, 6:53 am   #15
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

The LT1468 may be worth looking at. It's a little noisier than the NE5534 (5nV versus 3.5nV per root Hz) but very low bias current and oodles of bandwidth.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 25th Oct 2017, 7:29 am   #16
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Thanks RW. Yes much lower bias current than the LM833. On the downside, only a single amp so I would need to rejig the socket arrangement for two of them, but that shouldnt be too much trouble. Much appreciated tip.

Tim
TIMTAPE is offline  
Old 25th Oct 2017, 9:50 am   #17
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

There may be a dual version of it if you search LT's website.

For any transistor whether FET or bipolar, whether discrete or as an element in an IC, the noisiness varies with the impedance presented to it (both real [resistive] and imaginary [capacitive/inductive] terms) and with its bias current.

Some devices running very low currents give best noise behaviour with high impedance sources. Others running higher currents give their best noise performance with lower impedance sources.

In the RF world, you get contour maps of noise figure on Smith charts of the source impedance which could be applied.

In the opamp world you get a simplified version where they quote a noise current as well as a noise voltage. You have to look at the noise voltage, and also at the noise current multiplied by the source impedance. Sometimes the noise current does more damage.

The NE5534 although it has high bias currents, doesn't have quite as bad a noise current as might be expected. Its noise voltage is very good. Altogether the NE5534 is good with low-medium impedance sources. and apart from the problem of DC coupling its bias current into a tape head, it would otherwise be a good device to follow a head (if AC coupled!)

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 25th Oct 2017, 1:59 pm   #18
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Thanks again RW. As I understand it, on a reel to reel tape machine the noise performance is not normally critical as track width and tape speed are usually relatively large and high. It's more an issue with slow speed, narrow track tape formats like microcassettes, linear audio tracks on VHS, Beta etc where there's very little output from the tape head.
I suspect the issue here is the risk of tape magnetisation. Would decoupling the head with appropriately rated capacitors solve the problem?

Tim
TIMTAPE is offline  
Old 28th Oct 2017, 11:08 pm   #19
REDGUUZ
Triode
 
Join Date: May 2017
Location: Assen, Netherlands.
Posts: 13
Default Re: REVOX A700 Transport Control and Capstanboard: Underdesigned transistors?

Sorry to report back rather late.
In the meantime I have overhauled the capstan, tapecontrol, relay and power supply boards to avoid that a faulty tantalum capacitor would wreck an unobtainable IC (TCA561, TDA1000 or SC10429).
I have also replaced the BC108B drivers (following the TCA561's) by BC141-16 's (as in the STUDER B67Mk1). It looks like the Tapacontrol board was already designed for these transistors.

wrt the noise issue of the TBA931 Op Amp's: On my deck all TBA931/2 's had already been replaced by TBA931/3 by a former owner/service technician (apparently in 1991).

BTW There is a chap in Germany offering a drop in replacement for the TBA931/3 using the OPA2134 OP Amp see:
https://www.ebay.de/itm/Ersatz-TBA93...item2807bbc016

Any comments on this replacement?

best regards, Martin
Attached Thumbnails
Click image for larger version

Name:	Tapecontrolboard Voor.jpg
Views:	961
Size:	68.3 KB
ID:	151552   Click image for larger version

Name:	Tapecontrolboard Na.jpg
Views:	1221
Size:	84.2 KB
ID:	151553  
REDGUUZ is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 8:05 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.