Otari MX-5050 BII Problem - Not for the faint of heart!

[DanH]

David you have a very good point.. After all this troubleshooting with Leo and 2N, I know we've all but isolated the problem to somewhere in the motor/servo pcb section and I've learned a ton about this machine and electronics troubleshooting in general along the way. If these last few steps prove fruitless, I intend to further ravage my parts queen.

Now, if I replace the whole motor and have the same problem, things would get really interesting!

[DanH]

Alright, did some more probing. Pin 1 of IC1 shows the following:


You might not be able to see it in the pic, but that's + and -20V (the counter says 3.5V but its showing RMS). The positive to negative swing is about 2 seconds apart, while negative to positive is about 3 seconds apart.

Looking at the terminal for purple, which is connected to C1 and then on to IC1 pin 6, I found the following signal going from around +500mV to 2V very slowly:


When I saw this waveform coming off the tach pickup, it suddenly seemed to make sense that after replacing the motor I noticed playback speed fluctuating from fast to slow in an asymmetrical but repeating pattern. Am I right is concluding that this signal from the tach pickup is causing the motor to brake, then release brake, then brake again etc?

Yesterday I had the "brilliant" idea of hooking the two poles of the replacement tach pickup to the scope, to observe a signal when passing an electromagnetic field across it (hopefully, verifying its functionality). Unfortunately, the only EM field I could think to create was from my degausser. Is it possible I inadvertently magnetized the replacement head in a manner which would produce this result?

Thanks for your continued interest!

[LeeYoo]

Hi Dan.
The signal amplitude is ok, but the frequency is not.
So, the IC works correctly, but the pickup head is not.
Even if the motor had only one pole, you should already see the same frequency as the RPM of the motor.
??There might be 20 or so poles on that magnet??
I would expect maybe a hundred pules per second here. And a 20v t/t square wave on IC1,pin1.
??You might have magnetized the pickup head.??
Don't demagnetize this while connected. You might blow the IC's input.
Are those two little blocks attached to the head little magnets, or just mounting blocks?
You can test the pickup by connecting it to your scope, and rub a magnetized screwdriver over the front. ??Expect maybe 100mv t/t??.
Protect the head with a piece of paper.
Or rotate the motor by hand while the pickup is mounted.
Can you adjust the proximity of the head to the rotor?

Leo..

[DanH]

Originally Posted by LeeYoo

??You might have magnetized the pickup head.??
Or rotate the motor by hand while the pickup is mounted.
Can you adjust the proximity of the head to the rotor?

Yeah, I'm wondering if I did exactly that. I definitely took it out of circuit before testing it though. Those two little blocks appear to be magnets.

At one point I was seeing a full square wave on IC1 pin 1, something what looked like an on/off sort of signal. I had to play with triggering, coupling and resolution on the scope though because probing certain places started to load the motor down. I'll be honest - I don't know what "t/t" means!

The gap between the pickup head and the sawtooth gear wheel on the motor can be adjusted between 0mm (touching) and about 10mm away. Its set in place by those two small screws, which are visible through a cutout on the back of a fully assembled motor.

[LeeYoo]

Originally Posted by DanH

I don't know what "t/t" means!

Haha, just top/top.
The open collector output of IC1 can swing between the 24volt rail (R5 pull-up) and ground.
The IC amplifies about 1000x (R4/R3), so even a small input voltage from the pick-up head drives the output voltage to rail or to ground.
I did not see the teeth untill now. This means that the frequency is depending on the teeth passing the pickup.
Maybe your only problem is cleaning the surface of the pickup-head, and adjusting the gap till you have a proper 50/50 square wave on IC1,pin1.

Leo..

[David Kulka]

When all this is over you should check speed with the supply reel near empty, and also find some way of checking flutter. I remember that pinch roller solenoid pressure is critical on those machines and if the capstan is polished, that can cause problems. Also, as on most other machines, capstan motor azimuth errors can cause tape path problems - make sure the tape path is centered and stable across the heads and guides.

[DanH]

Well guys, I have some great news. The speed control now works!

I decided to dissect the original tach sensor, and discovered intermittent continuity between the outer connection posts and the coil windings on the inside. So that would be why the sensor wasn't sending any signal, or if it did, only for a few seconds.

The second sensor I put in the motor, I'm almost certain I permanently destroyed with the degausser. The sensor is bracketed by two small magnets - when testing, I should have used a magnetized screwdriver or kitchen magnet or something like Leo suggested. Instead, I used a degausser - and while I did see results, I probably permanently magnetized or demagnetized the coils inside and surrounding magnets. LESSON LEARNED.

So, I went to my third parts machine, took out the whole servo PCB and motor, stuck it in this machine, and voila! Speed control seems to respond properly (subject to further confirmation). Obviously I'll now need to realign the machine.

To everyone who contributed to this odyssey, THANK YOU!!! While I may have been able to solve this from the beginning by swapping the whole assembly, the knowledge I picked up along the way will prove invaluable as I take on fixing another two machines we have here. I can't thank you enough!

Hopefully, I won't be back on here in a few hours with another huge problem

Best,
Daniel

[LeeYoo]

Hi Dan.
Well done.
About the other sensor, I highly doubt that this passive component can be destroyed by a degaussing wand.
I also think, the field is also too weak to neutralise the magnets.
Is the coil resistance about the same as the good one?
Did you give that other pickup a try?
Leo..

[DanH]

Originally Posted by LeeYoo

Hi Dan.
Is the coil resistance about the same as the good one?
Did you give that other pickup a try?
Leo..

Hi Leo, thanks and thank you for all your help. From what I remember, the resistance on the first coil was around 3.2 kOhms, and the second coil was about 1.4 kOhms. I did put in the second pickup.. the capstan would not respond to speed selection changes, and instead played tape at a fluctuating speed resembling this waveform:



There's about 5mm of adjustment play in placement of the sensor inside the motor chassis. I tried to place it at approximately the same position that it was in before taking it out. You think its possible a change in spacing would change operation that much?

[LeeYoo]

Hi Dan.
I would have expected the coil resistance to be about 200ohm.
You said, you did fix the broken wire on the pickup.
A high resistance as you measured would be very, very fine wire.
As for adjusting: measure the output on the IC, pin1 with the scope.
Solder a wire to it, to bring it out of the machine if you can't reach.
A clean square wave is wat you want. Almost the full swing of the supply voltage.
Leo..

[DanH]

Indeed, those coils are made with very fine wire! I'll post a close up picture when I'm back at work.

To clarify, I did not fix the broken wire in the first coil. Doing so would have been extremely difficult since the short was on the inside of the coil's back cover, and there was no slack.

The square wave I'm looking for... I'll scope the now-functioning machine for its frequency and voltage. While this machine works now, there are several others here.. And since I robbed some of them for parts, I'm sure I'll be facing this problem again.

Thanks again - I can't believe how much I've learned from you guys on this odyssey!

[LeeYoo]

Hi Dan.
I think the coil wires break because of the mechanical load of the big wires soldered onto them.
Or maybe desolved from the chemicals in the solder flux.
If you can't get new coils, try fixing the old ones with liquid wire.
This is a silver glue, commonly used to fix rear window demisters in cars.
Leo..

[2N1305]

Congratulations Dan, you debugged to the lowest level, you truly are a serious tech! An open coil... I didn't think you could get that. Although guitar pickups can go open-circuit.

In any case, this was an excellent example of how to debug properly. I think this is one the best threads I've read!

Give this man a Gearslutz T-Shirt!

ps: as Leeyoo pointed out, maybe the flux was eating at the copper wire and that's what caused it to break, roughly twenty years later.

Cheers,
2N

[LeeYoo]

Originally Posted by 2N1305

Give this man a Gearslutz T-Shirt!

+1

Most slutz give up when they see the finish line. Or they just stop posting, so we all have learned nothing from it. Frustrating for the people who want to help them.
Not this one..

Leo..

[adkerche]

Hi, I have a MX-5050 MKIII-8 track with the exact same problem. My High/Low switch does seem to have some effect on the speed but it is certainly not doubling the speed. The pitch knob has no noticeable effect. I am getting about 1.44 khz off my 1khz MRL tone.

Anyway my main question is what machines could I pull a capstan motor and PCB from? Are all of the MX-5050's compatible? I know a guy who has several (6-10) of the BII's and BIII's. He does have a few of the MKIII's but they are the 2 or 4 track versions.

Thanks for everyone who was involved in this thread. I feel like I would be lost in the dark without it.

[DanH]

I've been swamped at work and havent been able to hang here recently, so sorry for the wait. Especially after Leo and 2n were so helpful to me!

As soon as I saw 1.44kHz, a look of dread came over me. Haha I don't know if the 8 track and 2 track 5050s have interchangeable capstan servo pcbs, but on there should be three little trimmers for speed calibration. Do they have any effect on playback speed?

This was a critical point of diagnosis during my saga and could help determine whether you really need to swap at all.

That said if you're in NY you're welcome to try one of my motors/pcbs..I have a few laying here.

Dan

Sent from my ADR6400L using Gearslutz.com

[DanH]

Also, see if you can upload some pics!

Sent from my ADR6400L using Gearslutz.com

[2N1305]

Dan, I hope you have no objection in me linking this thread to another Gearslutz member who seems to have a similar problem with his Otari deck?

I was reading the geekslutz threads (cause I'm a geek, or so I've been told) and saw this thread about a capstan motor going all wonky and wrong voltages everywhere... I suddenly remembered your thread!

Cheers,

2N

ps: If you ever see an old MX7308 like, I mean a really old Otari, 8 track, I'd be interested in taking it.

[rjhinton]

Dan,
I'm having trouble with the inductor coils in one of my MX5050BII decks and am considering replacing them with omnipolar hall effect sensors. What are your thoughts?

[DanH]

I love checking in on this thread after a while.. seems like this same type of problem keeps cropping up.

2N I certainly don't mind! I'd hope this saga would help at least someone. Besides, its really you guys who talked me through and did all the work. :D

Jr if you could provide a part number of what you'd replace with I could provide an informed opinion. In my motor there's about a 1/4" of play where that sensor goes..

If anyone's interested, the machine is still kicking ass, and I run it all the time.. will post some samples when I get home.
Sent from my ADR6400L using Gearslutz App

[DanH]

For those interested, here's a quick audio comparison. This is the live mix off our board, done while the performance was happening.. will happily detail the A/D conversion and signal flow if anyone cares.

RMG 911 tape, machine calibrated at 355nW/m, recorded +1.. nothing crazy, not slamming the tape hard.. subtle differences with this conservative use. I end up dumping the digital version and putting the tape bounce into the production queue, 9.5 times out of 10.

The file starts with the tape version, switches to digital at 1:05, and back to tape at 1:35. Pics of waveforms with amplitude stats on the right are attached. Cool band, will add the video once edits are done!

[DanH]

just realized GS crunches images.. here are proper links:

http://i.imgur.com/tZZy8.png (digital)

http://i.imgur.com/P0cuC.png (tape)