Marantz SD-160 Cassette Deck: Speed inconsistent between record and play

Thread Starter

Benjamin3000

Joined Mar 28, 2021
14
There is an error of about 7 Hz (confirmed with a frequency counter app) between the record and playback speed (recording is slightly too fast: cassettes recorded on other decks sound "in tune" on this deck when playing). As a result, music recorded on this deck sounds out of tune (flat) even when playing its own cassettes (and obviously, when playing those cassettes on other decks as well).

As mentioned above, the speed sounds dead-on when it is playing pre-recorded tapes, or ones recorded on my other machine. However, I'd like to get this one recording, since my other recorder, a GE boom box, does not have level meters or adjustments like this one does, so recording at the right volume is somewhat of a guessing game with the boom box.

The belt seemed tight enough on the pulley that I believe it is fine. My experience is that a bad belt would cause wow and flutter in addition to playing slowly.

Does anyone know, based on experience, how speed is supposed to be regulated in this machine? I don't believe I saw a sensor pointed at the "flywheel" if that's the right term (the metal wheel with the pulley, whose shaft is the capstan). My understanding is that some recorders have active feedback from a speed sensor of some kind, but simpler ones just have a voltage regulator. It would be a simple matter to find the adjustment for speed if it was consistently off by the same amount. However, the fact that it changes between record and play leads me to believe there's some kind of problem with speed regulation.
 

drjohsmith

Joined Dec 13, 2021
851
Uhm.
Typically , the speed regulation of these machines is built into the motor casing.
An amazing level of miniaturisation.
So I find it hard to understand why record and play would be different speeds .
Looking at the 7 Hz difference ,
How have you measured that.
Typically one would set speed by playing a calibration tape with a fixed tone on it.
Like 1 khz.
7 Hz off is going to be very very hard to distinguish .
I used to be involved in broadcasting.
If we played a 30 minute broadcast from big reel to reel tapes , we always allowed an extra minute , to allow for tape speed difference.
7 Hz is just to auditable on 1 khz.
Could the audio system of the machine be off.
Some capacitors dying ? Magnatised metalwork etc seem much more likely than speed .
 

Thread Starter

Benjamin3000

Joined Mar 28, 2021
14
Uhm.
Typically , the speed regulation of these machines is built into the motor casing.
An amazing level of miniaturisation.
So I find it hard to understand why record and play would be different speeds .
Looking at the 7 Hz difference ,
How have you measured that.
Typically one would set speed by playing a calibration tape with a fixed tone on it.
Like 1 khz.
7 Hz off is going to be very very hard to distinguish .
I used to be involved in broadcasting.
If we played a 30 minute broadcast from big reel to reel tapes , we always allowed an extra minute , to allow for tape speed difference.
7 Hz is just to auditable on 1 khz.
Could the audio system of the machine be off.
Some capacitors dying ? Magnatised metalwork etc seem much more likely than speed .

I measured it using an app on my phone. I played a 1khz tone into the recorder and then rewound and played it back, and put the frequency counter app on while the phone was next to the speakers.

I could tell the speed was off before then because the pitch of the music being played back wasn't quite right. Someone on a subreddit suggested using a frequency counter app to see if it was off and by how much. Tapes recorded in this deck also don't sound right in other decks either, but more confusingly the record speed seems off from play speed?

I have another player with adjustable play speed from the outside (with a center niche so you can tell where "normal" is for that player) and I can make the music sound right by adjusting that slightly fast, so it must be a speed issue.
 

MrSalts

Joined Apr 2, 2020
2,767
I'm guessing there is some drag on the recording head (dust, dirt, skin oil, peanut butter). Try cleaning with isopropyl alcohol and q-tip to clean any goo followed by a a soft white or pink rubber pencil eraser (to clean of any oxidized metal and second alcohol cleaning. Let it dry and try again.
 

drjohsmith

Joined Dec 13, 2021
851
These machines, the tape drags over the same heads in record and playback
So if speed was off , it would be off on record and play , they would sound fine on this machine .
The fft your phone uses , will have a long integration time me thinks
Wondering if your hearing wow / flutter caused by , as others have said the mechanics aging.
If it was speed , record of a tove and playback on same machine would be same.
Speed different for record and playback does sound very unlikely. .
To check speed.
Make a tone on another machine. Exactly 10 minutes long.
Play back on same machine and see if still 10 mins .
Then play back on suspect machine , and report back what the time is.
 

Thread Starter

Benjamin3000

Joined Mar 28, 2021
14
Schematic Excerpt.png

The motors have 4 terminals, presumably the "A" and "B" ones are for speed feedback. The reason there's four potentiometers is because the deck has a high-speed dubbing feature. It appears, based on the note on the schematic, that the bias to the transistors, is raised by just 0.4 volts to switch it to high speed (when the "dubbing" and "high speed" switches are both in)

I looked for electrical noise or voltage instability while in play and record (I was particularly interested in anything that changed between play and record, due to the nature of this rather strange issue)

One theory I had was that the bias signal was somehow "leaking" into the speed feedback circuit or its power source due to a bad filtering component or a bad ground. An oscilloscope put that idea to rest pretty quickly, however!

The "speed feedback signal" or whatever it was, was a DC level with a small amount, but still easily measured, ripple on it (with the oscilloscope) but given the lack of capacitors or inductors in the speed feedback circuit, I figured this was normal as there wasn't really a component whose failure would likely add ripple while still allowing the signal to be there.

In fact, I never did find the root cause: Since I don't need the high speed dubbing, and I figured that adjusting the high-speed dubbing pot might still have a slight effect, I fudged it by adjusting the high-speed dubbing speed pot off to one end (probably the highest resistance where it "loads" the circuit the least). The speed is now consistent between (normal speed) record and play. I actually don't quite understand what the two A and B terminals on the motor actually do. I suspect that I'm masking a still-present fault?
 

Dodgydave

Joined Jun 22, 2012
11,277
It looks like Q501 and Q531 are pnp transistors and in Normal mode they are turned on, so both the presets are across A&B , and when in High speed they are turned off disconnecting the other presets in parallel to increase the speed, so the speed is controlled by the resistance between the terminals A&B on the motor.

So lower resistance is slower speed and vice versa.. a simple check would be to open circuit the collector of one of the transistors, and adjust the speed, and see what happens when you short out the preset.
 
Last edited:
Top