CPC 464 tape - weird speed problem

[itneb]

Hi all,

Because this is my first post, i introduce myself quickly.
I'm a french (sorry if my english isn't very good!) part time cpcist since my childhood, rediscovering the joy of using a CPC from time to time.

The tape drive of my Schneider CPC464 has a weird speed problem since years which i never managed to resolve:

It performs well for some times then the speed starts to decrease progessively during some seconds to a point where loading errors appear, stay more or less at this speed a moment with 'wow' effects then returns to normal speed in a more or less aggressive way leading to read errors if not already appeared before.

During this 'speed problem' event the mechanism become a little noisier. If a press REW of FF during this event, the rewind/fast forward performs slower than normal too. Even without tape inside i recognize if the drive performs at accurate speed or not by earing the noise.
It is unpredictable when this problem is triggered. Sometimes it does not occurs during several hours, sometimes it is triggered very often. The duration of this event varies but is rarely longer than one minute.

I managed to capture the problem on video, The problem starts at ~2:40 and finishes à ~3:00:

Things i have already done (thank to other topics of cpcwiki):
- replaced the belt
- cleaned heads/capstan/pinch roller with qtip and isopropilic alcool
- used sand paper on pinch roller a little (this partially resolved some wow effect i have on long duration tapes but it is unrelated to this topic)
- tested a looot of tapes (short and long duration).
- rewind tape before play does not resolve the problem as it appear even without tapes inside the deck.
- tested with tapes locked inside the door or bypassing the door.

I can't figure which part of the mechanism is responsible. Can it be electrical ? the motor is dead? 

Thanks in advance.

[dlfrsilver]

Does it makes the sound on all tapes, or only on some ?

Because some crap cassettes can lead to that, i had some which were doing this on my very modern tape reader.

[itneb]

It does this on all the tapes (I have ~50 tapes)

If i swap the tape with another during this `speed decrease event`, the issue remains.

Furthermore, even without tape, by earing the noise made by the mechanism and by watching carefully the  `takeup reel` (? i'm not sure of the name of this i mean the right thing that turns when playing without tape) i can see that it doesn't turn at the right speed.

[dlfrsilver]

or the power is unregular in it's arrival in the motor.

Mhhh..... maybe again capacitors problem ?

[itneb]

I have really poor knowledges in electronics. Can this be checked using a multimeter (i have a really cheap one) ?

I have another cpc464 keyboard (an amstrad one) which is not in good condition, it has others tapes problems (pinch roller dead) but not suffers from this speed issue. Could i cross check with it or simply use it for spares ?

[Audronic]

@itneb


Spool the tape to the end.
Rewind tape, and then try again.
Some times the tapes get slightly folded over and scrape on the plastic.
and/or ONE very small drop of lubricant on the Capstan (Metal shaft that rotates) bearing ONLY ONE Small Drop.
Then clean the Capstan of any excess oil.


Good luck    Ray

[Velktron]

Other things to check:

Can you easily turn the tapes by hand? Are they too stiff or too tightly wound? This may indicate excessive takeup torque, due to a stiffening takeup spindle clutch.
Do the tapes leave some kind of sticky residue on the head (if light brown or white, it's a binder breakdown problem. Worse still, it's "contagious" between tapes, if the tape path is not cleaned thoroughly.
In the CPC tape mech, motion to the takeup spindle is transmitted via an idler. Is it worn/irregular?

[itneb]

Spool the tape to the end.
Rewind tape, and then try again.

I do it all the time, even with other decks (i have an old Toshiba walkman) but this don't solve the problem.

and/or ONE very small drop of lubricant on the Capstan (Metal shaft that rotates) bearing ONLY ONE Small Drop.
Then clean the Capstan of any excess oil.

I'll try it. What kind of oil/manufacturer can i use ?

Can you easily turn the tapes by hand? Are they too stiff or too tightly wound?

Yes they turn easily by hand (before and after use with cpc) and play nicely with the walkman.

Do the tapes leave some kind of sticky residue on the head

I have never found this kind of residue. The pinch roller was a slighty brilliant brown but i cleaned it.

In the CPC tape mech, motion to the takeup spindle is transmitted via an idler. Is it worn/irregular?

it's difficult for me to ascertain but i think it's ok. Here is a photo of the mechanism i took when i unskrewed it.

[attach=2]
I can unscrew it again to take others photos if this can help.

[Velktron]

I would like to see the underneath of the unit -exactly how motion is transmitted from the capstan/flywheel (which has a constant, controlled speed, at least that's the intention ) to the takeup spindle (which has a variable speed depending on the tape's position, and thus uses a slip clutch).

There are several methods this can achieved and there are numerous tape mech patents (e.g. contact idlers, permanent gearing, a secondary rubber belt, or direct/indirect drive by a secondary motor, in luxury tape decks) but the CPC464 service manual doesn't make it very clear which is actually used, so a photo of the underside of the tape mech would be useful, with emphasis on the critical details of how the takeup spindle is powered.

In any case, are those gears that appear on the upper side used during playback? In old boomboxes with a similar mech, those gears can stiffen, and require disassembly and some lubrication.

(EDIT: by looking at the pic carefully, it seems that during PLAYBACK, motion is transmitted from that rubber-coated idler in the middle of the mech, to the ribbed upper part of the right spindle, which then pushes on the lower portion through a felt clutch. In FAST FORWARD mode, motion is transferred to the LOWER part of the right spindle directly, using the the gear above the idler, which makes it spin harder and faster. Is this correct?)

As for lubrication in general: for the capstan bearing you can use a drop of any light machine oil (e.g. Singer sewing machine oil, 3-on-1 general purpose DIY oil etc.), the best technique for applying it, other than disassembling the unit and removing the capstand and flywheel, is to put the unit in playback/pause mode, and let a drop seep down into the bearing, by applying it directly to the spinning capstan.

For plastic, rubber or metal-on-plastic parts such as gears and spindles, things are a bit more complicated. You cannot use any mineral-based oils or grease there, because they break down plastic parts and cause worse problems on the long run. Only silicone-based lubricants are compatible with plastics, but they are harder to find.

Finally, another possibility to look into, is the motor itself. It also has bearings that can dry or wear out, causing speed fluctuations (usually works OK for a while but then loses speed suddenly and for a few seconds at a time, sometimes even squealing).

[itneb]

Wow! What an answer! Thanks Velktron

I'll take time this week to unskrew the tape mech, take photos/videos of the parts you asked and post them here.

usually works OK for a while but then loses speed suddenly and for a few seconds at a time, sometimes even squaealing

It sounds a lot like my issue and seems to be more accurate than a gear/dry issue because it is observed regardless of the 'mech position' (PLAY/REW/FF ...) but i may be wrong!

[Velktron]

If it happens in all modes then it's almost certainly a motor problem. Things to troubleshoot include the speed control potentiometer (if it's external to the motor):  it may need just some cleaning and working it about a few times (you will need to re-calibrate the speed or mark the old position, though). If the pot is built-in the motor though...that's gonna require a motor substitution. Luckily this kind of motor was common, and you might able to get it new or cannibalize it from an old boombox.

Another thing to look for is the motor mounts: usually, the motor is held on the frame by using some rubber shock-absorbing "feet". If those wear/rot out, the motor starts wobbling about, causing all sorts of problems.

[itneb]

it's almost certainly a motor problem

i came to this conclusion before posting on cpcwiki but having not a lot technical knowledges about tapes i prefered to ask experts!

If the pot is built-in the motor though...that's gonna require a motor substitution

I think it is the case because after watching photos of tape motors on the net i had not been able to find such control on the motor drive of my cpc. I even removed the black plastic surrounding the motor (the photo of the previous post was taken before) with the asumption it was under.

the motor is held on the frame by using some rubber shock-absorbing

I can't remember such thing but i'll take a look at this when i'll unskrew it.

I have another cpc464 with a good motor (at least the tape mech does not have this issue). I am ready to "cannibalise" it because it as other issues and is less usable than the Schneider.
But swapping motors is like a surgery for me because i'm not a soldering expert despite i have one soldering iron. In a previous post of this topic, @dlfrsilver mentionned a capacitor problem. Does it worth investigate in this way before ?

About lubricants on the gears, this product  is available in my local store. Does it fits ? :
SILICONE LUBRICANT, HIGH-PERFORMANCE LUBRICATION

Again thanks for your support!

[Velktron]

If the potentiometer is not easily found on the CPC's motherboard, then it must be built-in the motor, and only accessible though a small round hole at the back, for which you must use a very small jeweller's/watchmaker's screwdriver (preferably a plastic non-conductive one). It's hard to tell from the pic which of those dark spots at the back is a hole and which just a welding point/rust/dirt spot

Some motors do indeed have a small cap shorted across their poles, but that's more for noise filtering purposes. In addition, the motor used in the pic has no such cap (at least externally). It would help to take a pic of where the leads that power the motor are soldered on the motherboard.

As for the oil, yes, it's suitable, but avoid spraying it directly on the gears or anywhere else: try getting a few drops on a q-tip, and apply it indirectly. I personally dislike spray-on lubricants, as they tend to also get oil where you don't want them to, but silicon-based oil seems to be only available in spray form

[itneb]

If the potentiometer is not easily found on the CPC's motherboard.

I didn't know it could be on the motherboard. I was aware of a possible screw to adjust the speed on some motors after searching about this subject on the net but i didn't know it was mandatory.
One thing i'm sure there is no such screw on the motor. The spots you see on the pic are holes but not screws. They seemed to be sort of thin paper layer protecting the inside of the motor. 

It would help to take a pic of where the leads that power the motor are soldered on the motherboard

i'll do that when i'll unskrew it.

[Velktron]

One thing i'm sure there is no such screw on the motor. The spots you see on the pic are holes but not screws. They seemed to be sort of thin paper layer protecting the inside of the motor.

If the speed pot is on the motor, it will be inside the casing, protected by that paper-thin cover (usually rubber). But some motors have the holes without anything behind them. In any case, the service manual should reveal if the motor pot is on the motherboard or not. And yes, it's mandatory because speed control in a tape deck is necessary, not an option. Getting it "for the most part" right based just on the factory specs of the motor (which can be off by more than a few % between individual motors) is not enough, since most tape decks don't have a closed-loop speed control, but are simply set at the factory.

[Bryce]

The motor speed is fine. If it wasn't the problem would be permanent. Don't start messing with the speed, because you'll need an oscilloscope to get it back to where it should be.
The deck just needs a decent service. Replace the belt, clean the capstans and cogs with a dry toothbrush (I use the wifes ). Then slap some grease (vaseline works great) on the cogs teeth and between all moving parts, including the metal parts from the buttons (just don't get any on the capstans). There's also a few open microswitches, you should spray some contact cleaner on these. Don't try to clean them with any physical contact method, you'll bend them out of shape. Then clean/roughen the rubber pinch wheels with fine sand paper. Clean the heads with alcohol using a cotton bud (Coton-tige). That should get the deck back to normal.

Bryce.

[Velktron]

The motor speed is fine

Such a random variation of speed which affects all mechanical functions in all modes, is not exactly what I'd call "fine" 


Unless you mean it's "within normal parameters" 

[Bryce]

Such a random variation of speed which affects all mechanical functions in all modes, is not exactly what I'd call "fine" 


Unless you mean it's "within normal parameters" 

No of course it's not fine, but it starts at the correct speed and motor speed calibration won't stop variance. It would only change the start speed before the problem occurs. Same with a drive. If the calibration is correct and it varies, then it was something else causing the variance, not the calibration.

Bryce.

[itneb]

I'll go tomorrow to the store to buy the weapons (grease vaseline, contact cleaner , silicone lubricant etc. you mentionned) and will try to give it a decent service.

[itneb]

I've taken some photos

rear with pcb mounted:
[attach=2]

rear pcb unmounted:
[attach=3]

more to come...

[itneb]

Front view of the pcb:
[attach=2]

[itneb]

I have problems posting more photos triggering security checks. Maybe should i wait a little..

[itneb]

Here is the rear of the motor:
[attach=2]

[itneb]

The motor reference:
[attach=2]

mech cogs:
[attach=3]

[itneb]

Finally the heads:
[attach=2]