-
Feeling lucky? Check out the AK 24th Anniversary Giveaway thread and join in the fun!
Servicing a SONY DAT machine (Pictorial)
- Thread starter SaSi
- Start date
SaSi
Seriously Illogical
No, but I think this one calls for a total stripdown. It's a good opprtunity to document the whole thing.
The capstan seems to function OK, but I am not sure that there is no oil that leaked and contaminated the capstan's shaft.
In any case, I removed the capstan for cleaning. Although it is not necessary to remove the shaft itself, it helps both document and handle the two white washers.
To remove the capstan the 3 small cross screws need to be undone from the top of the transport. These are shown in the first picture.
The rotor is secured in the motor assembly by magnetic force. To further reduce the effects of stray cables inside the assembled transport it is shielded with a clear plastic cover. One screw secures it.
In the 5th and sixth pictures, the two white washers are clearly visible. They are there to protect the ends of the shaft from tape particles that could otherwise attach the friction points.
To remove the rotor only a gentle pull is needed. Also care is required so that the two white washers don't jump and get lost.
The two last pictures show the rotor and windings of the capstan motor.
I cleaned and reassembled the rotor / capstan shaft but it didn't make any difference (negtive or positive).
Judging by the behavior I see, I would guess the problem is combined gummed up lubricant in various places and a weak (?) solenoid (the one that disengages the breaks).
Attachments
SaSi
Seriously Illogical
OK, I compared the transport exhibiting the FF/REW/PLAY sluggish performance with one that operates fine and was able to track down the fault.
I need to refer you to a previous picture. We need to discuss the two solenoids.
The yellow one (horizontal orientation in the pic) is responsible for back tension. The black one is responsible for the breaks.
The difference that I noticed between the two transports in operation is that the "good" one will plunge the solenoid and keep it plunged while the tape is moving. As soon as transport is stopped the solenoid is released and the breaks engange.
The "faulty" transport seems to plunge the solenid at all the correct times, but the solenid doesn't keep position but instead is released. This results in the breaks being enganged while the tape should move. Not really easy to do that.
Now that the reason for that has been found, I need to find the reason why the plunger doesn't remain in place. It is obvious that for that to happen a steady current flow must be ensured through the plunger. It appears that is not the case.
I am studying the service manual to see what could be wrong. There is a large TO-220 transistor showin in the service manual as driving the plunger and I wonder what mode of failure can cause the transistor to deliver pulses but not keep them alive. There is also a buffer IC there used in several functions related with motors and solenoids. That could be also suspect.
I am now thinking of putting two decks on the bench. One that operates flawlessly and another with a problem. The fault appears not certain to be isolated on the transport mechanism. So, an exchange of transports between a faulty and a working deck appears tempting.
I need to refer you to a previous picture. We need to discuss the two solenoids.
The yellow one (horizontal orientation in the pic) is responsible for back tension. The black one is responsible for the breaks.
The difference that I noticed between the two transports in operation is that the "good" one will plunge the solenoid and keep it plunged while the tape is moving. As soon as transport is stopped the solenoid is released and the breaks engange.
The "faulty" transport seems to plunge the solenid at all the correct times, but the solenid doesn't keep position but instead is released. This results in the breaks being enganged while the tape should move. Not really easy to do that.
Now that the reason for that has been found, I need to find the reason why the plunger doesn't remain in place. It is obvious that for that to happen a steady current flow must be ensured through the plunger. It appears that is not the case.
I am studying the service manual to see what could be wrong. There is a large TO-220 transistor showin in the service manual as driving the plunger and I wonder what mode of failure can cause the transistor to deliver pulses but not keep them alive. There is also a buffer IC there used in several functions related with motors and solenoids. That could be also suspect.
I am now thinking of putting two decks on the bench. One that operates flawlessly and another with a problem. The fault appears not certain to be isolated on the transport mechanism. So, an exchange of transports between a faulty and a working deck appears tempting.
vincei
aka MasterControlMedia
Interesting, I wish you were around 10 years ago! I sold my MINT in box, low hours 59ES for next to nothing because of the sluggish transport. 
I seem to recall a service bulletin for these transports. In fact I recall trying to order a new complete transport and got a look like I was nuts from the guy at the counter. 
I seem to recall a service bulletin for these transports. In fact I recall trying to order a new complete transport and got a look like I was nuts from the guy at the counter.
SaSi
Seriously Illogical
Interesting, I wish you were around 10 years ago! I sold my MINT in box, low hours 59ES for next to nothing because of the sluggish transport.
Sorry to hear that vincei. But it is interesting. Low hours mint condition. The sluggish transport problem doesn't appear to be tied to 1000s of hours of operation or years of idling in storage then. Therefore it CAN be an electronic part failure.
Did you use 60 and 120 minute audio tapes only or did you also use DDS 90m and 120m tapes giving 3h and 4h play time?
The reason I'm asking is that these tapes have almost double the weight of tape compared to the normal audio tapes and that could cause marginal operation due to the higher load.
Torgue in the DAT drives are in the order of single digit g/cm for the SONY transports. Direct drive transports (like Panasonic and JVC) can have 50-150g/cm torgue but the Sony ones are marginal for heavyweight tapes - hence the problem with moaning FF/REW and failure to rewind 120m tapes even in perfectly running machines.
vincei
aka MasterControlMedia
I always used audio DATS from Sony or Maxell. I used 60, 90 and 120 minute tapes, all had the same problem. I had only put about 200 hrs on the unit at most, maybe lack of use was a problem. I bought it in the fall of 1992 and it started having the problem about 1998. I seem to recall looking at the transport at the time and also suspecting the solenoid wasn't releasing the brakes enough. I also bought a used A7 DAT that had a completly frozen takeup reel.
SaSi
Seriously Illogical
Lack of use should be the reason for failure on heavyweight heavily lubricated and greased machines. Although I have seen some grease gumming in sony DAT transports (1 out of 15 units), I think that these sparcely lubricated machines don't fail if not used (save for some atmospheric contamination on the head drum that might render the unit mute (until cleaned).
SaSi
Seriously Illogical
Good News and Bad News
At the title says:
I brought up from the basement two different units, reason being I had to compare two identical machines, one having the fault and one not. And the two easiest to access machines were 2 DTC ZE700 units. Actually one, as I already had a problem unit on the bench...
I verified that the transport on the "good" machine operated flawlessly and then I transplanted it to the host deck that originally housed the "faulty" mechanism. It doesn't operate any more. Not even play. BUT it DOES rewind the tape back in the supply reel when eject is pressed (like all "faulty" transports do).
I then put the both transports back to the host deck that originally hosted the "good" transport. The one originally there of course operates flawlessly. The second one is also working flawlessly.
Therefore:
The sluggish FF/REW and sometimes PLAY fault is NOT a fault of the transport but a fault in the main PCB that hosts the control electronics for motors and solenoids. Most likely the fault is lack of steady signal to the break solenoid. The breaks need to be released during tape transport and the fault is failure to release the brakes.
Now some bad news:
Firstly, my attempts to fix the problem transport included heavy cleaning and removal of the slightly sticky tension pads. The result is that the "previously-thought-of-faulty" transport operates freely now, but a tad too freely - no back tension. The result is that FF/REW is free but a bit jerky and lack of back tension causes the tape to curl off the pinch roller.
Secondly, as I considered how to work on the main PCB I realized removal of that PCB is a major task. Not difficult or risky of braking things but big time tedious.
Thankfully, I have a set of DTC ZE700 electronics in a box removed from an enclosure that was shot. The electronics operate ok and I hope the PCB has the same problem so I can work on that and identify the fault.
A hint: Touching the power transistor on the faulty PCB I feel it hot and I also feel it rather loose on it's leads. Like something is unsoldered or broken underneath. On the good machine, several cycles of FF/REW don't seem to increase temperature to a level I can feel it by touching.
Now that the fault is isolated on the main PCB I have to work towards fixing the "faulty" transports using the good deck as a testing unit. My plan is to clean and lubricate the transports, replace loose and missing tension pads and document that procedure. When this is over, I will work on the PCB faults that cause the transport failure.
The reason is: All these units share a virtually identical transport but differ in the electronics. I also feel more comfortable with the mechanical aspects rather than the electronic ones (for now). Therefore as soon as transport issues are eliminated and described, we can delve into the unknown and troubleshoot the circuits.
At the title says:
I brought up from the basement two different units, reason being I had to compare two identical machines, one having the fault and one not. And the two easiest to access machines were 2 DTC ZE700 units. Actually one, as I already had a problem unit on the bench...
I verified that the transport on the "good" machine operated flawlessly and then I transplanted it to the host deck that originally housed the "faulty" mechanism. It doesn't operate any more. Not even play. BUT it DOES rewind the tape back in the supply reel when eject is pressed (like all "faulty" transports do).
I then put the both transports back to the host deck that originally hosted the "good" transport. The one originally there of course operates flawlessly. The second one is also working flawlessly.
Therefore:
The sluggish FF/REW and sometimes PLAY fault is NOT a fault of the transport but a fault in the main PCB that hosts the control electronics for motors and solenoids. Most likely the fault is lack of steady signal to the break solenoid. The breaks need to be released during tape transport and the fault is failure to release the brakes.
Now some bad news:
Firstly, my attempts to fix the problem transport included heavy cleaning and removal of the slightly sticky tension pads. The result is that the "previously-thought-of-faulty" transport operates freely now, but a tad too freely - no back tension. The result is that FF/REW is free but a bit jerky and lack of back tension causes the tape to curl off the pinch roller.
Secondly, as I considered how to work on the main PCB I realized removal of that PCB is a major task. Not difficult or risky of braking things but big time tedious.
Thankfully, I have a set of DTC ZE700 electronics in a box removed from an enclosure that was shot. The electronics operate ok and I hope the PCB has the same problem so I can work on that and identify the fault.
A hint: Touching the power transistor on the faulty PCB I feel it hot and I also feel it rather loose on it's leads. Like something is unsoldered or broken underneath. On the good machine, several cycles of FF/REW don't seem to increase temperature to a level I can feel it by touching.
Now that the fault is isolated on the main PCB I have to work towards fixing the "faulty" transports using the good deck as a testing unit. My plan is to clean and lubricate the transports, replace loose and missing tension pads and document that procedure. When this is over, I will work on the PCB faults that cause the transport failure.
The reason is: All these units share a virtually identical transport but differ in the electronics. I also feel more comfortable with the mechanical aspects rather than the electronic ones (for now). Therefore as soon as transport issues are eliminated and described, we can delve into the unknown and troubleshoot the circuits.
SaSi
Seriously Illogical
Will most certainly try resoldering first.
SaSi
Seriously Illogical
Further tests
Well, I brought out the electronics parts from that shot ZE700 and laid them on the bench.
The good news is that both transports (the verified one and the previously suspect one) operate in the same way they did with the good host.
The bad news is that I cannot troubleshoot this PCB :sigh: That means that I need to remove the PCB from the other deck and use that instead.
That's a considerable amount of work and pictures and for now I only took a few pictures of the "workbench" version of the deck. More will come tomorrow evening.
For those who might have lost this, I've temporarily stopped working on the DTC-59ES as I had to compare working and non working transports, which later moved to comparing working and non working PCBs - as the transports were proven inoccent.
I have quite a few DAT units around, but unfortunatelly only one DTC59ES, therefore troubleshooting that fault needs to switch into troubleshooting DTC ZE700s of which I have plenty.
As a sidenote, the ZE700 is a much newer deck which adds the SBM circuit (Super Bit Mapping). Apart from that, it appears to be more or less a DTC690, which appears to be a cut down version of the DTC59ES.
The circuitry in all these machines are quite similar, even the designation of certain parts is the same. I've compared the schematics of the 59 and the ZE700 and the spool motor / plunger driver circuits are quite similar. Therefore, I am hoping that solving the problem on the ZE700 will point directly to the offender on the DTC59 as well.
I hope the above cleared things rather than confusing the issue even further.
Well, I brought out the electronics parts from that shot ZE700 and laid them on the bench.
The good news is that both transports (the verified one and the previously suspect one) operate in the same way they did with the good host.
The bad news is that I cannot troubleshoot this PCB :sigh: That means that I need to remove the PCB from the other deck and use that instead.
That's a considerable amount of work and pictures and for now I only took a few pictures of the "workbench" version of the deck. More will come tomorrow evening.
For those who might have lost this, I've temporarily stopped working on the DTC-59ES as I had to compare working and non working transports, which later moved to comparing working and non working PCBs - as the transports were proven inoccent.
I have quite a few DAT units around, but unfortunatelly only one DTC59ES, therefore troubleshooting that fault needs to switch into troubleshooting DTC ZE700s of which I have plenty.
As a sidenote, the ZE700 is a much newer deck which adds the SBM circuit (Super Bit Mapping). Apart from that, it appears to be more or less a DTC690, which appears to be a cut down version of the DTC59ES.
The circuitry in all these machines are quite similar, even the designation of certain parts is the same. I've compared the schematics of the 59 and the ZE700 and the spool motor / plunger driver circuits are quite similar. Therefore, I am hoping that solving the problem on the ZE700 will point directly to the offender on the DTC59 as well.
I hope the above cleared things rather than confusing the issue even further.
Attachments
SaSi
Seriously Illogical
Getting Closer
I removed a faulty PCB from a deck. The pictures show the particular unit and the red circles show where plastic retainers need to be twisted with pliers. In total 8 white retainers are to be released.
In addition to the retainers, two plugs from the transformer and another two from the headphone amp and line in volume control need to be disconnected.
Finally, 4 brass screws need to be undone in the back, close to the analog and digital inputs and outputs.
Removal of the PCB isn't as hard as I recalled for this deck. It was hard on the particular one I worked on as that was crushed and deformed and I had to work around so that the PCB wouldn't break. Strangely that PCB not only survived but also performed well in the "sluggish transport" department.
I removed a faulty PCB from a deck. The pictures show the particular unit and the red circles show where plastic retainers need to be twisted with pliers. In total 8 white retainers are to be released.
In addition to the retainers, two plugs from the transformer and another two from the headphone amp and line in volume control need to be disconnected.
Finally, 4 brass screws need to be undone in the back, close to the analog and digital inputs and outputs.
Removal of the PCB isn't as hard as I recalled for this deck. It was hard on the particular one I worked on as that was crushed and deformed and I had to work around so that the PCB wouldn't break. Strangely that PCB not only survived but also performed well in the "sluggish transport" department.
Attachments
SaSi
Seriously Illogical
The main PCB
The next few pictures show the main PCB from the underside, mainly to show a peek on the ICs. This particular deck (DTC-ZE700) has the ICs surface mounted at the bottom.
By the number of jumpers on the board, I assume it is a single layer board. Therefore the SMDs need to go under. So much about high-end and expensive...
The second picture shows a general view of the motor and solenoid control area from below. The third picture is a close up of Q451, the TO220 transistor I felt being rather loose. The solder is broken there. Fortunatelly the tracks are intact.
The first thing I tried was to resolder that particular transistor (although it is not the one responsible for driving the offending solenoid).
As I worked on resoldering that and a few other joints I saw as suspect, I was in for a big surprise. Quite a few of the soldering pads, when warmed by the soldering iron, lose their solder. The last picture attached shows Q459 and the middle lead shows how solder is going away (onto the soldering iron).
It appears that the PCB has very little (I dare say inadequte) solder on the tracks.
I re-soldered all the passive components near the motor and solenoid control area and also the transistors.
The next few pictures show the main PCB from the underside, mainly to show a peek on the ICs. This particular deck (DTC-ZE700) has the ICs surface mounted at the bottom.
By the number of jumpers on the board, I assume it is a single layer board. Therefore the SMDs need to go under. So much about high-end and expensive...
The second picture shows a general view of the motor and solenoid control area from below. The third picture is a close up of Q451, the TO220 transistor I felt being rather loose. The solder is broken there. Fortunatelly the tracks are intact.
The first thing I tried was to resolder that particular transistor (although it is not the one responsible for driving the offending solenoid).
As I worked on resoldering that and a few other joints I saw as suspect, I was in for a big surprise. Quite a few of the soldering pads, when warmed by the soldering iron, lose their solder. The last picture attached shows Q459 and the middle lead shows how solder is going away (onto the soldering iron).
It appears that the PCB has very little (I dare say inadequte) solder on the tracks.
I re-soldered all the passive components near the motor and solenoid control area and also the transistors.
Attachments
SaSi
Seriously Illogical
Retesting
As soon as this was finished, I reconnected that faulty PCB back to the testbed version of a DAT and tried 3 mechanisms I have handy. None of them operated properly and with all of them the same problem. The brake solenoid doesn't release.
I attach a block diagram of the solenoid driver and a shot of the actual circuit driving it. It is driven by Q481.
The service manual diagrams are useful as they provide reference voltages to check across the circuit. So I ventured to probe the referenced voltages. More or less the readings are what the circuit describes.
I measured 3.54 at the base and 4.65 at the collector while the emmiter voltage was just a bit lower than 2.8V. Compared with the voltages measured on a good-working board, these values are within the measurement error.
By measurement error I mean that these values are there for 1-2 seconds after PLAY or FF/REW is pressed. Soon after that the deck decides it can't move the tape and aborts.
I was not surprised to measure these figures. I was surprised to see these values last that long while the solenoid just fires and retracts. I was expecting to barely measure any pulses as I was thinking that for some reason, instead of providing a steady current, the circuit was merely producing a brief pulse.
I did measurements both in peak hold mode and in RMS mode. The measurements don't differ.
Now, I am really stuck. I can't figure out why this is happening. The circuit appears to be healthy, all solenoids measure approximately the same resistance (19-20Ohm), yet some PCBs are able to hold the solenoids in while others can't (the same solenoids). :scratch2::scratch2::scratch2:
Now would be a good time for any kind of suggestions.
As soon as this was finished, I reconnected that faulty PCB back to the testbed version of a DAT and tried 3 mechanisms I have handy. None of them operated properly and with all of them the same problem. The brake solenoid doesn't release.
I attach a block diagram of the solenoid driver and a shot of the actual circuit driving it. It is driven by Q481.
The service manual diagrams are useful as they provide reference voltages to check across the circuit. So I ventured to probe the referenced voltages. More or less the readings are what the circuit describes.
I measured 3.54 at the base and 4.65 at the collector while the emmiter voltage was just a bit lower than 2.8V. Compared with the voltages measured on a good-working board, these values are within the measurement error.
By measurement error I mean that these values are there for 1-2 seconds after PLAY or FF/REW is pressed. Soon after that the deck decides it can't move the tape and aborts.
I was not surprised to measure these figures. I was surprised to see these values last that long while the solenoid just fires and retracts. I was expecting to barely measure any pulses as I was thinking that for some reason, instead of providing a steady current, the circuit was merely producing a brief pulse.
I did measurements both in peak hold mode and in RMS mode. The measurements don't differ.
Now, I am really stuck. I can't figure out why this is happening. The circuit appears to be healthy, all solenoids measure approximately the same resistance (19-20Ohm), yet some PCBs are able to hold the solenoids in while others can't (the same solenoids). :scratch2::scratch2::scratch2:
Now would be a good time for any kind of suggestions.
Attachments
SaSi
Seriously Illogical
B+ is the positive power rail for most circuits. There appear to be a couple of decoupling capacitors near the driving circuit (10nF) plus a medium electrolytic I need to check where exactly sits.
"Solid ground return": Combined with the previous point, do you suspect lack of adequate current available and / or increased resistance reducing the driving current below the level that would keep the solenoid retracted?
Is is possible that a circuit unable to deliver enough current would deliver a peak (resulting in the solenoid kicking in) but not sustain it to keep the solenoid kicked in?
It does make sense to me. I am thinking that I should measure the voltage across the solenoid socket. Whatever I measure at the circuit needs to travel halfway across the PCB, go through the white ribbon cable to the drum drive board and then through the socket to the solenoid. Would it make sense to bypass all this and connect the solenoid directly to Q481?
I cannot suspect whatever exists from the ribbon cable to the solenoid as the fault remains with the PCB rather than follow the transport assembly. So, I am focusing from the ribbon socket to the LSI giving the initial pulse. I am still trying to find a proper spot to probe for the pulse. I think I need to switch to a scope - the DMM doesn't help that much in transient pulses.
I am also thinking to put the recent findings into use and check the failure mode back on the DTC-59. The circuitry is much less dense and somewhat neater in layout.
"Solid ground return": Combined with the previous point, do you suspect lack of adequate current available and / or increased resistance reducing the driving current below the level that would keep the solenoid retracted?
Is is possible that a circuit unable to deliver enough current would deliver a peak (resulting in the solenoid kicking in) but not sustain it to keep the solenoid kicked in?
It does make sense to me. I am thinking that I should measure the voltage across the solenoid socket. Whatever I measure at the circuit needs to travel halfway across the PCB, go through the white ribbon cable to the drum drive board and then through the socket to the solenoid. Would it make sense to bypass all this and connect the solenoid directly to Q481?
I cannot suspect whatever exists from the ribbon cable to the solenoid as the fault remains with the PCB rather than follow the transport assembly. So, I am focusing from the ribbon socket to the LSI giving the initial pulse. I am still trying to find a proper spot to probe for the pulse. I think I need to switch to a scope - the DMM doesn't help that much in transient pulses.
I am also thinking to put the recent findings into use and check the failure mode back on the DTC-59. The circuitry is much less dense and somewhat neater in layout.
Tinman
Super Member
Well, it was a known issue when I worked in a shop that sony stuff of that era had a lot of bad solder. We did re-solder a lot of boards. That fixed many weird faults.
The white ribbons were replaced as well, though I still don't know what made them go bad.
The Capacitor in the B+ is a guess. I had a similar fault on a linear tracker TT. The tonearm solenoid would engage, but not stay locked. I bridged across the main cap for that supply.... problem solved. I replaced the cap.
The point is, I repaired so many "dogs" that I don't rule out anything anymore.
But... a quick test is to bridge the cap, and if that doesn't do it, run a wire from the transistor right to the solenoid connector to test the board. You could have a resistive solder connection. Yes, I have had those. Especially with the transitional junk when they went to lead free solder.
It's gotten better, but I still hate the new crap. Very difficult to work with.
Speaking of working with stuff, I am restoring an old STUDER mixing console. Talk about a DREAM to work on. EVERYTHING comes apart. Even the sliders are made to be serviced. Man, oh man... what quality. It's a pleasure to restore.
We'll NEVER see that sort of engineering again.
The white ribbons were replaced as well, though I still don't know what made them go bad.
The Capacitor in the B+ is a guess. I had a similar fault on a linear tracker TT. The tonearm solenoid would engage, but not stay locked. I bridged across the main cap for that supply.... problem solved. I replaced the cap.
The point is, I repaired so many "dogs" that I don't rule out anything anymore.
But... a quick test is to bridge the cap, and if that doesn't do it, run a wire from the transistor right to the solenoid connector to test the board. You could have a resistive solder connection. Yes, I have had those. Especially with the transitional junk when they went to lead free solder.
It's gotten better, but I still hate the new crap. Very difficult to work with.
Speaking of working with stuff, I am restoring an old STUDER mixing console. Talk about a DREAM to work on. EVERYTHING comes apart. Even the sliders are made to be serviced. Man, oh man... what quality. It's a pleasure to restore.
We'll NEVER see that sort of engineering again.
Thought on Solenoid Problem
I am new to DAT decks and obtained one is a bizarre way. I was in an old recording studio and observed a Sony Deck that I thought was a cassette deck. I asked if it was a cassette deck and the owner of the studio said it was a DAT deck. This deck was at the bottom of a pile of other old audio equipment and he said I could have it if wanted it. It was a rack mount unit with a cassette tape stuck in it, but is really good shape considering. It powered up, but that was it. I gave me a strange error on the display. I opened it up and found that a tiny belt had jumped off a pulley. I replaced it and the deck works now. I am an old analogue cassette user and this thing is way too complicated for me. I am trying to find an operators manual. It is a Sony DTC-8A, but that does not bring up any information when I Google for it. It does record, but there are a lot of features I don't understand.
Anyway, on the old cassette decks that used solenoids, there was a configuration that provides full power to the solenoid to engage it, but this produces a lot of heat within the solenoid. A circuit was designed that would fire the solenoid (usually a capacitor/transistor combination) when the solenoid was engaged the voltage was dropped to just hold it against the load. I have seen conditions when the solenoid would not fire, but if you "helped it" it would hold the load, and then I have seen the condition you have described with the solenoid firing but dropping out.
I see a lot of IC's in my unit and the solenoid function may be handled by one of the IC's. On the old cassette decks it was easier to fix since it was more oriented to individual components.
Sorry about the long explaination. It may not even be revelant.
I am new to DAT decks and obtained one is a bizarre way. I was in an old recording studio and observed a Sony Deck that I thought was a cassette deck. I asked if it was a cassette deck and the owner of the studio said it was a DAT deck. This deck was at the bottom of a pile of other old audio equipment and he said I could have it if wanted it. It was a rack mount unit with a cassette tape stuck in it, but is really good shape considering. It powered up, but that was it. I gave me a strange error on the display. I opened it up and found that a tiny belt had jumped off a pulley. I replaced it and the deck works now. I am an old analogue cassette user and this thing is way too complicated for me. I am trying to find an operators manual. It is a Sony DTC-8A, but that does not bring up any information when I Google for it. It does record, but there are a lot of features I don't understand.
Anyway, on the old cassette decks that used solenoids, there was a configuration that provides full power to the solenoid to engage it, but this produces a lot of heat within the solenoid. A circuit was designed that would fire the solenoid (usually a capacitor/transistor combination) when the solenoid was engaged the voltage was dropped to just hold it against the load. I have seen conditions when the solenoid would not fire, but if you "helped it" it would hold the load, and then I have seen the condition you have described with the solenoid firing but dropping out.
I see a lot of IC's in my unit and the solenoid function may be handled by one of the IC's. On the old cassette decks it was easier to fix since it was more oriented to individual components.
Sorry about the long explaination. It may not even be revelant.