Tascam 58-OB Story...

I suppose "oddball" wasn't really the right term to use...oddball for me since I don't have them! :D

Just need 1 each 100uF 10V polar, 22uF 16V bipolar and 4.7uF 50V polar 105 degree caps. They'll get here...not like I don't have OTHER stuff I can do in the meantime. :p
 
I checked my stash of Panasonic FC's and 105 degree Nichicons, but I do not have those values...I do have a bunch of 1000uf 6.3 volts however! :D
 
Tearing through some of the smaller projects...

Recapped one of my DX-4D's that will be mated with the 58 last night.

Reminders: no, I don't think everything must be recapped, but it is what I'm working toward. There may or may not be any noticeable change in performance with a full recap. My experience? My M-520 was a whole lot quieter after the PSU was recapped so it can make a difference. Overall it is an insurance policy. They do get old (capacitors), and that can cause trouble. Eventually one is bound to go. So its just something I'm doing and it is getting faster...though I still have tails to snip, inspecting to do and a little cleanup and reassembly, it only took me about three hours to do what is pictured below, and that included some wasted time:

The schematic that I have has a number of inconsistencies in cap values with what was installed at the factory so I had to go through the unit with a fine tooth comb to get that all documented...if a schematic calls for a higher value than what was installed I'll generally go with what is on the schematic. If it is the other way around and the factory installed anything with a higher value than what the schematic calls for I defer to what was installed. All the inconsistencies involved higher values being used by the factory. So that took some hunt and peck time...plus I have a fabulous system for organizing all my capacitors :rolleyes:; its called throw all the little packages in a box...takes a little time to dig through and find everything I need...I think I resist getting a little organizer unit because I keep telling myself that, for the most part, the recapping will slow down extensively at some point and I would ask that you all stop chuckling now.

Here is a before shot of the DX-4D innards:

IMG_0754_1_1.JPG



Here is the "after but in-progress" shot:

IMG_0756_2_1.JPG



Here is a closeup of the encode/decode PCB which is where most of the caps are:

IMG_0757_3_1.JPG
 
Hi Cory,

How in the world will you acurately recalibrate the DBX ?

Regards,

Danny
 
Sheesh...didn't even consider that. I mean, I still think its a good thing to recap it, but if doing so requires recalibrating it then I guess that means I buy a service manual for the thing and learn how to calibrate it...Actually I can probably get some info on the process from my 234 service manual. The dbx PCB's between the DX-4D and the 234 are very similar.

But the way I see it, I figue I just calibrated it by recapping it. I haven't measured any of the caps I pulled out of the thing...Don't have an ESR meter anyway, but just in terms of value drift I figure they are pushing if not outside spec...so if that is true then the unit would need to be calibrated to compensate for the drift if that's even possible...if it was calibrated for in-spec caps, and they were out-of-spec, I just put it back the way it was at the factory...so...hm.

Anybody else wanna weigh in on this?

I have the calibration instructions for my dbx 150X's, and its not rocket science...can't be that much differen with the DX-4D but I'm thinking I don't even need to go there.
 
I finally got the rest of the 58 recapped tonight with the exception of the amp cards...I had to wait for quite awhile to get the last couple caps to do the Capstan Servo PCB. So that's done.

Now, I was getting ready to put the tach roller back on. I'd taken it off to check the bearings and it needs to go on before the Capstan Servo PCB gets remounted as the mount screw for the roller gets covered up by the PCB. Well, I took a close look at the rubber on the tach roller and SHOOT its bumpy and scaly...It ould probably be fine since its just an impedance roller but it would really stink if it ended up being a problem because that tach roller is basically about the deepest thing in a 58 to remove as the Capstan Servo PCB has to come out and all kinds of stuff has to come out before that comes out...I'm going to check the roller on my parts deck and see if it looks better, but then I've got to find a special size allen wrench to get the damaged cap off of that parts unit...I have 0.7mm and 0.9mm and the set screws that fix the cap on the roller assembly are in between that. :eek:

Here's a closeup of the tach roller:

IMG_2032_3_1.JPG



So once I figure that out then its simply time to put it back together...er...huh! How does it go back together?? Seriously, its been almost too long and things are getting fuzzy as far as how this goes back together...and then I need to prepare myself that it may not even function properly once its together since it is unknown at this point whether or not the damage has been repaired.

IMG_2034_4_1.JPG
 
Cory, that tach roller doesn't look bad. Lookit, it's basically dried out somewhat (it's certainly not gooey) and, if I were you, I'd do a once over on it with some rubber conditioner. I did just that on rollers which had that same appearance and it worked out fine. As you know, if it was gooey / sticky, then that's another story but it doesn't appear to be that and is a prime candidate for a rubber cleaner. Just use a lint free rag (cotton or microfibre) and make sure you "pull" a tiny bit off that black off to expose the good surface. Should be as good as new. :)
 
Oh, cool Daniel...Thanks!

I remember now that the original pinch roller looked like that and because the surface had a definite texture to it I opted to replace it, but that is a more critical roller...I have some conditioner/cleaner made by MG Chemical. I'll give that a try. ;)

Good news as a backup plan, I figured out how to disassemble the tach roller assembly and it can all come out the front leaving just the spindle which is what gets covered up by the Capstan servo PCB. So even if the conditioner doesn't work and I need to send it to Terry Witt, or if ever I need to replace the bearings in the assembly it won't require completely opening the patient up. So indeed I'll try the conditioner and if it is all good then I'm good for a time...BTW the bearings function like new...this thing really hasn't seen a lot of mileage.

As a tengental though I may just send my spare roller to Terry to have on hand for someday...I'll keep it with the spare low mileage complete headblock, new guides, blahblahblah :rolleyes:

Appreciating the tape path on this thing...Everything the tape touches is rolling on ball bearings except the heads and the stationary guides in the headblock...guess that's not really special per se but the rolling components are pretty beefy for 1/2 inch tape. Also, tape flutter is a by product of the total wrap (the sum of the angles the tape makes) around stationary components in the path...very little wrap on the 58 tape path, and then the other thing that is important is the total length of unsupported tape...less seems to be better...the length of unsupported tape drives the resonant frequency of the flutter...less unsupported tape would mean a higher flutter frequency placing it further out of reach of human ears...outside of closed loop systems the tape on a 58 is pretty well supported. And its cool that there aren't any impedance rollers...just less stuff in the tape path.
 
Good news/bad news

As per usual, I'm glutting a single post with a truckload of info, but I have a question and a need at the bottom of the post...if you read nothing else can you look at that and if you can help let me know?

The 58 is coming back together nicely. Night before last I got all the PCB's reinstalled (except of course for the amp cards...leaving those out until everything else works, and then per evm1024's advice from long ago testing them one at a time...). I was honestly nervous about it because as I started to piece it together I realized how much my memory had faded as to what went where, which screws fastened what...that sort of thing. Fortunately for the most part I labeled plugs well, re-threaded screws into brackets and parts so that it was obvious what went to what, and had pictures to reference. The manual is, of course, essential too to double and tiple-check things. Looks like this now:

IMG_2076_1_1.JPG


IMG_2078_2_1.JPG


IMG_2082_3_1.JPG



All the transport functions (except for one vital piece; more on that below...) operate wonderfully...all lights light up, reel motors are working nice and strong, solenoids are crisp...forgot how much I like the mechanical sounds of this machine. I'm really pleased with all that. I also confirmed how the tach roller comes apart and this would apply to the 52 as well:

You need a 0.9mm hex key. I thought it was too big and my 0.7mm too small, but it is just a snug fit for the 0.9mm...may just be the thckness of the plating on my 0.9mm bit or the manufacturing tolerance but it worked. I have those bits as well as others in a neat mini-bit driver kit I got in a set a year or two ago and it has been a real boon to my toolbox. So there are two of those 0.9mm hex socket set screws in the roller cap (the one that has the dimple in it for manual scrolling with your finger), and once those are loosened the cap comes right off and what you see inside is a philips screw in the center holding a thick brass washer on. It goes together like the pinch roller...loosen the screw and remove it and the washer and behind that the first bearing gets pushed out by the spring that is in between the front and rear bearing. In order to get the roller off now you'll need to remove the tach sensor. You can get to the two screws that fasten the tach sensor to the deck plate...they are to the left of the sensor. If you don't remove the sensor it will restrict the removal of the roller because of the slotted sensor plate on the back of the tach roller assembly. Now you can pull the roller assembly off. It is possible that you may not be able to get the rear bearing to come with the roller assembly. This is what happened to me. The rear bearing was on the roller spindle so snugly that it stayed on the spindle, and bearing OD is a tight squeeze through the ID slotted sensor plate. My sensor plate got slightly distorted but this shouldn't effect the performance of the tachometer circuit. I then had to use my little custom puller I made for reel motor bearings to get the rear bearing off the tach spindle. Its not press fit or anything, I just think, again, it might be a tolerance thing as it was really easy to get off using the puller. Usually if a bearing is press fit onto a spindle it will give a little *pop* when it comes free under the pressure of the puller. No *pop* at all here. So here are the parts:

IMG_2083_4_1.JPG


So now I'll probably replace the bearings and try cleaning up the rubber a bit with conditioner as cjacek suggested.

So now for the bad news:

Something is gunny-sack with my capstan motor or on the servo board...:( Used to be that when you powered the deck up the capstan motor gave this neat "brrrrrup" sound and quickly reached speed and then was nice and quiet. Now it goes "brrrrrrrrrrrrrrrrr", is spinning well below the correct speed if it spins up at all (sometimes I have to give it a spin with my fingers to kick start it). The pitch control function works. There is a funny clicking sound too when it reaches a certain speed. I think I used to hear that clicking when the capstan was under external sync control and when it was speeding up or down to lock to the timecode...

Here is a video of the problem so you can hear the sound (again, this is DivX encoded):

https://www.torridheatstudios.com/ftp/share/movies/Tascam%2058-OB%20Movies/MVI_2081-10.avi

Also, here is a pdf of the Capstan Servo PCB:

https://www.torridheatstudios.com/ftp/share/documentation/Tascam/Tascam%2058/Tascam%2058%20Capstan%20Servo%20PCB.pdf

I did check the voltages coming to the PCB and I think they are okay, but a couple are high:

PIN 1: 4.9V (should be 5.0V...A-OK)
PIN 2: GND (A-OK)
PIN 6: 29V (should be regulated 24V...I think that's okay...:confused:)
PIN 7: 34.5V (should be unregulated 24V...again, is that okay? Too High? :confused:)

I did recap the PCB so maybe that has something to do with it, or maybe one of the local regulators for this PCB got damaged when I put the extender card in wrong. That error did take out a 15V regulator in the PSU, so maybe something got damaged...dunno. Ideas?
 
I would expect unregulated 24 v to be 3 to 5 volts above 24 volts. Say 27 to 30 or so. 34 might be OK. but might be high. It is unregulated after all.

I would expect regulated 24 to be 24 (plus or minus .6 volts or so.).

I thinik that you have a bad regulator somewhere. You might also put your scope on the unregulated 24 line with the vert set to AC coupling and about 5 volts a divi. You should not see any significant ripple. I would expect less than 100 mV AC ripple (perhaps mush less). In any case if you have lost of ripple (volts or more for example) then you have a blown diode or few.

--Ethan
 
Actually the more I think about it the more I think that it is blown beyond repair.

Just finish cleaning it up and I'll take it off your hands....

Looking good in other words.

--Ethan
 
Just soiled my trousers...

All I saw of your last post, Ethan, was the first line...initially...and then I scrolled dwn and saw the rest, though I was in some sort of ill euphoric state so it took me about three reads to realize that what I should say now is something like "thanks but bite me" or something. :D

Going out to mess with it.

Thanks, Ethan.

Heart rate normal now.
 
Ethan,

Check out post 269.

I just read from there up through #283. When I made the tragic amp card offset error the capstan was still spinning fine. A few posts later I listed the voltages and the regulated 24V supply was running about 27.5V andthe unregulated supply was running 31.6. Obviously it is higher than that now, but even in post #283 I mention again that the capstan was spinning fine.

None of this may matter but I just thought I would mention it since the only things I did to the PSU PCB is replace the bad 15V 7815 regulator and recap it...which...I guess is a lot.

I just wanted to redirect to the original issue as there are likely clues in there to what may be happening now.

Going to try and find time to look for ripple in the PSU rails...pull each supply connector off the PCB's to isolate the PSU as much as I can. Prbably would have been a good idea to shotun the rectifiers on the PSU PCB...
 
Check all connectors (can they go on backwards?) Every thing where it should be?

Regulators should regulate.

--ethan
 
All connections are on the standard multi-pin connectors used in Teac gear...not possible to connect them offset or backwards (unless you really try hard I suppose...)

I isolated the power supply last night by disconnecting everything connected to the PSU (at least as much as you can...the JOINT PCB's are still hardwired but those connect to the solenoids and reel motors, and they won't do anything because the Control PCB is disconnected from the PSU).

I really think there is something failing in the capstan motor or on the Capstan Servo PCB...the problem is getting progressively worse. I scoped the +5V, +24V regulated and +24V unregulated rails...pictures below, but after I took those pictures I again connected the servo PCB to the PSU and now the motor doesn't turn at all unless you spin it by hand...that kick starts it...it actually has been that way the past couple times but each time it is spinning slower and slower when it reaches a stable speed. I'm guesstimating that last night it wasn't spinning fast enough to pull tape to even 3 3/4ips. Plus its really weak. Would be easy to stop it. I don't really know any technical stuff about this but I've read a couple threads over the last couple years where somebody is having a capstan problem and somebody has suggested the drive transistors...could this be a cause of my problem?

Here are the three pictures of the 3 above mentioned rails:

+5V
IMG_2104_3_1.JPG


+24V (regulated)
IMG_2105_4_1.JPG


+24V (unregulated)
IMG_2106_5_1.JPG


I don't get it...they all look the same...in all cases the CH1 VOLTS/DIV was set to 5, and CH1 was set to AC...the TIME/DIV was set to 5 and the probe was a 10x probe.

When I measured the same rails with a basic DMM I am still seeing +4.9V, +29V and +34.5V.

Time to pull the Servo PCB out and start shotgunning components or pull the PSU PCB out and start replacing the rest of the regulators and rectifiers? Or...both??? I'm up for whatever. This deck is going to work.
 
Make that 50 volts a division

With 10x probes that would be 50 volts / division. You are seeing 150 v peak to peak. I wold say that the scope does not have a common ground with the PS.

With AC coupling you are blocking the DC component so you will never see it (+5 etc).

So hook the probes ground to the PS ground.

You should only see mV of ripple on the unregulated and less than that on the regulated rails.

You should measure with some load as well.

--Ethan
 
Well, there's a reason that Ethan has over a million rep points. Not that that is news to anybody, but the high rep is warranted: he took some significant time this eve outside of the thread to give me a crash-course on the scope operation and we also dialogued about the 58.

I threw him for a loop because I thought I was using the 10x probe...it was the 1x.

Thing is, I still don't quite get why all the traces look the same even though the voltages on the three rails tested are considerably different...? :???:

I tested them again tonight armed with a better knowledge of how to trigger the scope and it was the same thing.

So the plan is, at this moment, for me to put together a test load out of a couple automotive 12V bulbs in series to see how the two 24V rails respond under load ; check the voltage drop...make sure the PSU is supplying good current. And I still don't think I've finished determining if the power coming on those rails is clean...

The schematic isn't clear at all about the wires that go from the PCB to the motor..I think there are about 10 of them but there is NO detail about their nature or purpose. Ethan has offered to look at the schematic.

So there will be more to report in the coming days.

Kudos to Ethan. Glad you are active on the forum and willing to help in this way when able. Several times I would be without hope were it not for your time and expertise. Thanks for your instruction! ;
 
Last edited:
You must log in or register to reply here.
Back
Top