Philips N4307 only records for a short time

[frsimen]

I assume that the record level meter is indicating that the signal you are recording is at the correct level throughout. You say that the playback fades away, rather than abruptly stopping. That doesn't sound like a poor joint/cracked PCB track or a switch problem.

Try changing C756, which may be faulty.

If that doesn't do the trick, what test gear do you have? Knowing that, it will be easier to provide further help.

Paula

[ghpille]

Originally Posted by frsimen

I assume that the record level meter is indicating that the signal you are recording is at the correct level throughout. You say that the playback fades away, rather than abruptly stopping. That doesn't sound like a poor joint/cracked PCB track or a switch problem.

Try changing C756, which may be faulty.

If that doesn't do the trick, what test gear do you have? Knowing that, it will be easier to provide further help.

Paula

Wouldn't a faulty C756 affect what I hear through the N4307's speaker?
I have a couple of simple voltmeters and a Red Pitaya, a credit-card sized computer with an FPGA, which can act as an oscilloscope. Someone on Youtube claims these things are worthless, better buy a 2nd hand real one. There's one for sale a couple of miles from here for 95€. I can probably get two working N4307's for that price, but where's the fun in that?

Your last message was posted at Today 12:39am, which seems to be earlier than Today 11:07am. Wow!

[julie_m]

Originally Posted by ghpille

Your last message was posted at Today 12:39am, which seems to be earlier than Today 11:07am. Wow!

"12:39 am" is 0:39 in VCR notation.

[ghpille]

So I've "tested" capacitors 756 and 733, with an ohm-meter. With 756 the resistance keeps slowly climbing as long as I measure, which seems to be normal behaviour for a working capacitor. But I left the capacitor in circuit, perhaps that invalidates my test?
I left the 733 also in circuit, but with track switch on PAR, so that one side of the capacitor is cut of from any possible short. I measured immediately an infinite resistance. Is the ohm meter capable of measuring a 500V capacitor? When under current, and recording, 10V arrives at that capacitor, of which 1mV reached the bias trim potentiometers.
I was reading about bias on Wikipedia. Is the N4307 using DC bias?
Shall I desolder the capacitors and test them as it should be done?

[ghpille]

Originally Posted by julie_m

"12:39 am" is 0:39 in VCR notation.

Thanks, Julie. I realised that before I started typing, but I still wanted to register my first reaction. With a career in IT I have been exposed to the anglican time and date notations, but they can still surprise me. 6/2/2025 comes after 2/6/2025.

But what is VCR notation?

[Uncle Bulgaria]

Capacitors must be tested out of circuit for reasonable readings unaffected by other circuit components. The only worthwhile test that can be done with a digital multimeter is to see whether they are short-circuit. Your test of infinite resistance is only made at the test voltage of the multimeter's battery, which is too low to be useful. A useful leakage test must be performed at their rated voltage, for which you need to rig up a power supply and ammeter to see whether the capacitor is passing excessive current at its rated voltage. IF you search for a capacitor leakage tester on the forum you will find many commercial and DIY possibilities. I use a homemade tester that's good for 630V, built from a magazine PCB, and a hand-cranked Megger insulation tester for anything higher.

What are the capacitor types? The only ones with a reasonable possibility of failure are electrolytics, which could well have failed with age. Ceramics, micas and film are very reliable and only very rarely the cause of a problem.

Have you inspected the PCB with a magnifying glass for any dry joints? They can be very hard to spot. It's tempting to break out the soldering iron and replace components, but it can easily cause more damage so it's worthwhile doing some thorough tests for continuity and a visual inspection.

As far as I'm aware, VCR is only Video Cassette Recorder. Your dates look American to me - in the UK 2/6/25 comes after 6/2/25.

[frsimen]

If C756 is leaking it won't make much difference to what you hear from the speaker as there is another capacitor between the output of the record amplifier and the input to the output amplifier. A leaky capacitor could be applying DC to the record/replay head and that could be increasing with time. I suggested replacing the capacitor to remove that as a possible problem.

The N4307 uses AC bias, running at 60kHz, there should be no DC reaching the head at all.

The voltages present in this circuit are a lot lower than 630V, so there is no need to test at that level! It's very unlikely that there is a problem with the capacitors in the erase oscillator circuit, especially as Sideband has explained that the one I was concerned about is a ceramic type and reliable.

There are several different versions of the Red Pitaya. The SDR version is not suitable for testing the recorder, the others are. The best thing to do is to get the Red Pitaya working as an oscilloscope. Connect it via a 100nF capacitor to the right hand side (on the circuit diagram - the side away from the head connection) of R565 and watch what happens as you make the recording. If the level drops as the recording progresses, the problem lies between there and the output of the amplifier. If it doesn't change, measure the DC voltage at that point using your meter. It should be zero. If it is higher than a few mV, there is DC coming from somewhere.

If the recording still fades, despite the level at R565 being the same, perhaps R565 is faulty.

You will probably need a X10 probe to measure on the other side of R565, due to the bias voltage.

The other thing to check is the voltage at MP1, across R528, or MP2, across R527 (depending on which track you have selected). It should be about 10-25mV rms (around 28-70mV p-p).

Again watch if anything changes as a recording is made.

Paula

[ghpille]

Originally Posted by Uncle Bulgaria

A useful leakage test must be performed at their rated voltage
What are the capacitor types?
Have you inspected the PCB with a magnifying glass for any dry joints? They can be very hard to spot. It's tempting to break out the soldering iron and replace components, but it can easily cause more damage so it's worthwhile doing some thorough tests for continuity and a visual inspection.

I've attached a picture of the C733. I have trouble reading the colour code. I get brown, black and yellow, which should be 10x100, ie. 1K (picoFarad?). Yellow for tolerance but then black which would be 100V? The schema says 1K 500V.
I did a continuity test, see above, and just now a visual.
I'll see about a DIY capacitor tester.

[frsimen]

The colour code on the capacitor is brown black red yellow.

That is correct for a 1000pF capacitor, the yellow represents the voltage rating, presumably 500V. I doubt it's faulty though, so don't get too fixated on this.

Paula

[Sideband]

That is a 'pin-up' capacitor that was mentioned earlier. Yes 1000pF ceramic and usually very reliable. I've only ever had one fail in my 50 years of servicing/restoration and that was in a TV line timebase. I doubt that it's faulty in this position.

[ghpille]

I got the RedPitaya running, which was an adventure in se, but had already found an old analog oscilloscope for 95€ (they can be found for half that price but this one was at 10' driving).

I started recording a 1000Hz sine wave and followed the signal from C756 (where it was alive and well) but lost it after R565. That resistor measured correctly 8K2 ohms. I added half a pinhead of fresh solder on each side of the resistor and started recording. But only on the first track. I found no other problem, so I lowered the read/write head and was able to record on both tracks.

The volume of the recording is somewhat quiet, but the sound is perfect, although during recording I have quite some static. That static is not on the tape.

Thank you all!

[ghpille]

Does this in any way correspond to the fact that I was able to record for a while?

[saxmaniac]

Any reason for being fixated on capacitors? In the absence of a methodical strategy, how about squirting freezer spray along all components in the record signal path and see if the fault is thermal? That's semiconductors, resistors etc not just caps. Flex the PCB in case that might just find a broken track or bad joint?

[ghpille]

Why would I do that? The problem was solved by resoldering the resistor. Two messages earlier than yours. I haven't been accused of lacking method before. I'm afraid that in my first test, touching the resistor while measuring it, was enough to repair the path temporarily.

[Uncle Bulgaria]

If you have the ability to record without the previous recording problem, now is the time to go through the service checks and ensure the signal levels and voltage readings match the manual and circuit diagram. You may find other faults or out of tolerance components.

Language is a problem when it comes to describing phenomena. 'Static' implies tape hiss to me, which could be at a normal level or could be a biasing error such that the machine is not set properly for the tape you are using. Without hearing it, we can't really help, so any hard evidence of voltages at test points or pictures of an oscilloscope screen are very helpful.

Well done with finding the resistor problem!

[ghpille]

Thanks, Uncle. The voltage checks were the first thing I did, since I didn't have enough experience reading a schema to understand the path I had to follow, what was the amplifier, even if the amplifier was involved. But every measure was quite close to the specs.

Got a new challenge now, a Braun TG 1020/4.