Diagnosing tape loading issues

Discussions about Sinclair ZX80 and ZX81 Hardware
User avatar
mrtinb
Posts: 1906
Joined: Fri Nov 06, 2015 5:44 pm
Location: Denmark
Contact:

Re: Diagnosing tape loading issues

Post by mrtinb »

The home built TZXduino I have, has no amplification. It just outputs 0V or 5V on the output pin to the ZX81 or Lambda 8300, and that works. If I measure the output from the TZXduino when playing a P-file, I get 2.5V which is the average of 0V and 5V.

What is the average voltage on your TZXduino, when you play a P-file?
Martin
https://zx.rtin.be
ZX81, Lambda 8300, Commodore 64, Mac G4 Cube
patters
Posts: 122
Joined: Thu Mar 11, 2021 12:34 pm

Re: Diagnosing tape loading issues

Post by patters »

Seems to be around 1.8-1.9V which would be around 4V of amplitude. Presumably we need to compare current too though. Mine does have the LM386 amplifier as Moggy mentioned.
patters
Posts: 122
Joined: Thu Mar 11, 2021 12:34 pm

Re: Diagnosing tape loading issues

Post by patters »

I just moved the working WESPI from the Issue 1 into the problem Issue 3 and it doesn't work in there. Same deal as for the Maxduino. The screen shows a desynced grey pattern that doesn't wobble and dance like the working Issue 1 while loading. It's as if EAR is dead. The bad part is that the ULA is not socketed.

I do notice that all the build pics for the WESPI are on an Issue 1. Coincidence, or is the ULA sensitivity known to be worse on an Issue 3?
User avatar
mrtinb
Posts: 1906
Joined: Fri Nov 06, 2015 5:44 pm
Location: Denmark
Contact:

Re: Diagnosing tape loading issues

Post by mrtinb »

I only have issue 3 PCBs, so that will ruin that theory. :D
Martin
https://zx.rtin.be
ZX81, Lambda 8300, Commodore 64, Mac G4 Cube
User avatar
1024MAK
Posts: 5103
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: Diagnosing tape loading issues

Post by 1024MAK »

I don’t recall any significant difference between the Ear and Mic circuitry between the issue one and issue three boards.
And due to the small value of the capacitors and the high value series resistors, it’s unlikely that plugging in the 9V DC into either will do any damage. Ripple not really being an issue, as the PSU would not be under load, hence the ripple voltage would be low in the mV area.

There will be slight differences between different issue ULA chips, and indeed between different ULAs of the same issue, due to manufacturing tolerances. Remember, the ULA was only ever intended to be used for digital circuits. Not analogue / audio use. This is the reason why they are rather ‘deaf’. For the machine to see a high, you have to get above the logic one threshold.

There may be a very slight drift in performance of a semiconductor over time, but this is normally insignificant.

Ceramic capacitors are solid, so they can’t leak. However some have a wax like coating. This may be what you have.

I can’t remember what actual voltage levels the ZX81 needs on it’s ear input. But the tape input is more primitive compared to a ZX Spectrum. Sinclair took notice of the difficulty with loading in the Zeddy, so used an improved input circuit in the ZX Spectrum ULA (a hysteresis circuit so that it could cope a bit better with electrical noise). Although it still needs a large voltage input for it to load. But do keep in mind that the ZX81 ignores the negative part of any input.

You may already have checked this, but are the Ear socket contacts making a good connection to the 3.5mm mono plug (you are using a mono plug aren’t you)? And are the contacts clean?

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
User avatar
1024MAK
Posts: 5103
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: Diagnosing tape loading issues

Post by 1024MAK »

patters wrote: Sat Apr 03, 2021 11:29 pm Seems to be around 1.8-1.9V which would be around 4V of amplitude. Presumably we need to compare current too though. Mine does have the LM386 amplifier as Moggy mentioned.
The current is a function of the voltage and of the source impedance and of the load impedance. So if you measured the voltage while everything was connected, that’s a valid measurement. But if you tested while the load (the ZX81) was not connected, then the figure is not a lot of use.

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
patters
Posts: 122
Joined: Thu Mar 11, 2021 12:34 pm

Re: Diagnosing tape loading issues

Post by patters »

Happy Easter chaps. The socket contacts grip much better on the non working Issue 3 than on the very loose working Issue 1. And loading also fails with the WESPI that was soldered in, which would appear to rule out connection integrity. I did also try plugging and unplugging many times in case there was an oxide layer. I am indeed using mono leads, the original black and grey ones, and I also tested another really short one which I got with the Maxduino.

My previous voltage measurement was across the unplugged jack, so will be invalid. While loading a game on the working Issue 1 I measured around 1.45V AC across the socket contacts on the Maxduino PCB. That suggests an amplitude of around 3V and I believe somewhere on this forum it was mentioned it needs to hit at least 2V.

Searching on the forum did read a thread in which you mentioned that the capacitor value would affect input sensitivity. And I found a separate web page which stated that ceramic capacitors can drift with age. Is it worth replacing C10? Or bypassing it even as a test? Can a ULA partly fail, or are they generally all or nothing?
Last edited by patters on Sun Apr 04, 2021 11:48 am, edited 3 times in total.
User avatar
mrtinb
Posts: 1906
Joined: Fri Nov 06, 2015 5:44 pm
Location: Denmark
Contact:

Re: Diagnosing tape loading issues

Post by mrtinb »

Since the ULA is an IC which is a circuit shrunken down, I believe part of an IC can be broken.
Martin
https://zx.rtin.be
ZX81, Lambda 8300, Commodore 64, Mac G4 Cube
User avatar
1024MAK
Posts: 5103
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: Diagnosing tape loading issues

Post by 1024MAK »

mrtinb wrote: Sun Apr 04, 2021 11:43 am Since the ULA is an IC which is a circuit shrunken down, I believe part of an IC can be broken.
Partial failure is certainly possible. However, I don’t believe that failure of the Ear/tape input part of the ULA is common, as it’s a low power system.

Ceramic capacitors are rather poor at actually being anywhere close to their specified value even when new, hence they often have a tolerance of 20%. Although the more expensive types can be better at 5%. Their capacitance also changes with temperature and even depends on the voltage across them. Although their capacitance gradually decreases with age, they don’t generally loose a significant amount due to age though compared to the wide tolerance they started off with.

But as they are cheap, sure, it’s worthwhile renewing it. I don’t recommend bypassing it, as then there will be no DC blocking, and that will definitely cause problems.

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: Diagnosing tape loading issues

Post by Moggy »

There is something you can do to increase the output of the 386,if it hasn't been done already by the makers and will take a bit of confidence on your part to try.

Like most op-amps used for audio they rely on a bias circuit to control the strength of the out-put.
The 386 is a bit unusual in that it already has a small bias built in which gives it an amplification of X20.
The addition of a 10uf electrolytic cap' can boost this to X200 (the chips max out-put) and would certainly put to bed any thoughts of the out-put being too low.

I don't have a device to check so can you see if there are any tracks/components/resistors etc that are connecting pins 1-8 together of the 386. If none are present then the chip is working at the lower end thump wise.

The diagram attached shows the principle. forget about anything else on the diagram just focus on pins 1-8.

I personally have no qualms about experimenting but understand if you are more cautious and in this example I would remove the 386 from its socket and put it back in along with the 10uf cap' leads as a friction fit just for testing purposes, but again understand any caution you may have
Attachments
daYpX.jpg
Post Reply