Very low signal on MIC port

Discussions about Sinclair ZX80 and ZX81 Hardware
100monkeys
Posts: 6
Joined: Sat Sep 20, 2008 7:33 pm

Very low signal on MIC port

Post by 100monkeys » Tue Sep 23, 2008 10:25 am

Hi.
I had 2 ZX81's that both had very very low output signal. Hardly noticeable and probably leakage as the little signal disapears when i disconnect the UHF out.
I get the stripes as usual during save.

I see on diagram c12 47 Pf (pico) and c11 47 Nf (nano)
Both caps are ok.

Could it be a typo and should these be switched?
47 Pf seems low value for audio signal?

(Meanwhile i blew the ULA during measuring of one of the ZX's :oops: )

User avatar
yerzmyey
Posts: 1239
Joined: Thu May 15, 2008 10:11 am
Location: Rubber Planet
Contact:

Re: Very low signal on MIC port

Post by yerzmyey » Tue Sep 23, 2008 1:37 pm

Yo,
there was similar topic made by me, some time ago.
I've got 2 machines too. Both has the same feature. Therefore I know it is a typical feature of ZX81's contruction. A very low level of MIC output.
Which is most annoying as musical demos also use this connector (grrrrrr).
I will perform additional experiments in some time (I'm a lazy bastard, I realize that) and then I'll write something new about the subject-matter.
There must be some soldering/electronic way to boost the out-signal but it's not my area of expertise.
Sadly.
IN NIHILUM REVERTERIS - a big text-adventure game for ZX81: http://tiny.pl/g2m6m
"MONOCHROME" issue 5 - (Spring 2014) free paper/PDF magazine about ZX81: http://tiny.pl/q2m44
ZX81 COMPETITIONS 2007/2009: http://zx81.republika.pl/

100monkeys
Posts: 6
Joined: Sat Sep 20, 2008 7:33 pm

Re: Very low signal on MIC port

Post by 100monkeys » Wed Sep 24, 2008 1:59 am

yerzmyey wrote:Yo,
there was similar topic made by me, some time ago.
I've got 2 machines too. Both has the same feature. Therefore I know it is a typical feature of ZX81's contruction. A very low level of MIC output.
Which is most annoying as musical demos also use this connector (grrrrrr).
I will perform additional experiments in some time (I'm a lazy bastard, I realize that) and then I'll write something new about the subject-matter.
There must be some soldering/electronic way to boost the out-signal but it's not my area of expertise.
Sadly.
Thanks for clearing that up.
I thought It would be kinda wierd that my presumed typo did not show in google search. So it must be something else.
I still do not get the idea of the output filter, but thats just me then.
The signal is not just very low, but near to absent, and, as luck has it, i probably had 2 boards with the same fault?

As i said, 1 died, atleast i think the ULA could not handle my measuring probing.
Pin 16 now only does Vsync/Hsync and no video at all.
I did not try blind typing to see if the rest still works
Alas another dead ZX81 I'm afraid :(
I still own an Issue 1 board, and i'll try to be more careful to test further.
Boosting just gets me more background noise, the signal is just not there, but at pin 16 it IS!. So must be a PCB fault i guess.

Regards

User avatar
yerzmyey
Posts: 1239
Joined: Thu May 15, 2008 10:11 am
Location: Rubber Planet
Contact:

Re: Very low signal on MIC port

Post by yerzmyey » Wed Sep 24, 2008 9:43 am

100monkeys:

>The signal is not just very low, but near to absent,
----------
Yes, I would tend to agree.
And it comes a little bit stronger when I use extremly short cable (less of losts).
That's why at first I thought I have broken ZX-81.
Therefore I bought another one and... still the same. So that must be simply original output-level from the comp.
Supposedly in past they were using different tape-recorders, with much higher level of input on their MIC ports.
Now we use mostly line-in, or even simply PC computers to save progs. Usual audio-out from various machines is much stronger so also audio-in levels can be lower. It might cause mentioned problem: to low level of contemporary machines' audio-in.
Btw, the same goes to TV-out. I thought my comp was broken but guy from the Forum informed me that contemporary TV-sets usually don't work with ZX-81 becasuse its TV-output is also far too low for them.
I had to buy another ('tourist') TV-set and it works fine with ZX-81.


>and, as luck has it, i probably had 2 boards with the same fault?
---------
Oh, it would make 4 of them (including 2 of mine) so I don't think so.



About the hardware/soldering stuff, I hope somebody here will be able to help You, as I'm not a hardware-guy, You know. ;)
IN NIHILUM REVERTERIS - a big text-adventure game for ZX81: http://tiny.pl/g2m6m
"MONOCHROME" issue 5 - (Spring 2014) free paper/PDF magazine about ZX81: http://tiny.pl/q2m44
ZX81 COMPETITIONS 2007/2009: http://zx81.republika.pl/

User avatar
kpalser
Posts: 57
Joined: Sun Jun 03, 2012 2:18 pm
Location: Dundee, Scotland

Re: Very low signal on MIC port

Post by kpalser » Mon Dec 16, 2013 11:12 pm

In the event that my evening of experimenting might be useful to others googling this and subsequent similar threads ...

My three ZX81 (like everyone else’s unmoded units) are producing a weak signal from the MIC port. Since I wanted to double check the documented:
Each pulse is split into a 150us High period, and 150us Low period
I plugged it into my MacBook Pro in line port intending to record a sample. Unsurprisingly it didn't register at all due to the signal strength. My impulsive solution was to pop down to Maplin and buy their cheap SP-24B preamp.
zoomedout.png
in the centre a 1 bit (9 high pulses) with 0 bits (4 high pulses) to left and right
zoomedout.png (111.14 KiB) Viewed 1428 times
zoomedin.png
300us pulse length (150us high and 150us low)
zoomedin.png (66.21 KiB) Viewed 1428 times
I used the highest sampling resolution possible by my laptop, 384kHz, to see visualise the square wave. Again no surprises - pulses have a 300us length.

One point that did surprise me was how poorly defined the square waves are. Is this just a general characteristic of ZX80/ZX81s and perhaps similar era microcomputers? Or is it down to my preamp, cables and recording hardware?

gozzo
Posts: 448
Joined: Fri Jul 08, 2011 8:52 pm

Re: Very low signal on MIC port

Post by gozzo » Tue Dec 17, 2013 12:44 am

Try removing C11 (0.047uf) (or changing it to lower value) from the zx81 board, may 'improve' wave shape as this is from the MIC output to ground... to increase the ouput, reduce the value of R29 (1M)...but may need to also increase the value proportionately of C12 (47pf) as well...

User avatar
1024MAK
Posts: 2438
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: Very low signal on MIC port

Post by 1024MAK » Tue Dec 17, 2013 2:43 am

The signal from the ULA was intentionally heavily attenuated, because the mono tape recorders of the day were designed to work with low cost, high sensitively microphones.

Make sure that all connectors are very clean and making good contact. With such a low level signal, the slightest bit of muck, or poor connection can make a big difference.

If you don't want to modify your ZX81, try using a battery powered microphone preamp.

Mark

Edited 'cus I said "low sensitively microphones" when I meant "high sensitively microphones" :oops:
Last edited by 1024MAK on Sat Dec 21, 2013 10:27 pm, edited 1 time in total.

User avatar
PokeMon
Posts: 2225
Joined: Sat Sep 17, 2011 6:48 pm

Re: Very low signal on MIC port

Post by PokeMon » Tue Dec 17, 2013 2:54 pm

In the ZX80CORE I modified the attenuation.
This was done for very sensitive microphones of the 80ies or like the Phono input for some crystal phonographs.
This is very uncommon today and very fault-prone.
If you use an amplifier, you amplify the noise too.

They attenuated the signal by factor 1000 - so 5V signal will be reduced to 5 mV.
I changed R29 to 100k and C12 to 470pF (just devided/multiplied by 10) for not to change the characteristics but increase the level to acceptable 50mV.
This can be recorded with normal sound cards of the PC.

And by the way, the documentation of the nocash emubase is correct.
I used it to write a LOAD routine for the ZX-IDE to send .p files via soundcard to the ZX81 or ZX80 or to just create WAVs from them.
And the timing is not very critical as it is as slow as a snake. :mrgreen:

One member of the germany board (b0d0) wrote a loader using a simple printer port LPT to create the signals for the ZX81.
The internal load routine in fact just accumulates the overall time of data to detect a high or low bit.
So a high is just double length of a low and the quantity of pulses are used to have at least a frequency which could be handled/recorded by usual cassette players/recorders.

User avatar
kpalser
Posts: 57
Joined: Sun Jun 03, 2012 2:18 pm
Location: Dundee, Scotland

Re: Very low signal on MIC port

Post by kpalser » Sat Dec 21, 2013 8:31 pm

Thanks for the information and suggestions.

A little bit of background - I’m just trying to get wave characteristics that accurately represents the original device, flaws and all, for an emulator project. For the moment I’m holding off making any mods to the device. I did follow Mark's suggestion of using a battery powered pre-amp. The following screenshots show a recording under the same conditions but with a series of batteries powering the same preamp and with the laptop recording its output also unplugged from the mains for good measure.
battery1.png
Sample using battery powered pre-amp: in the centre a 1 bit (9 high pulses) with 0 bits (4 high pulses) to left and right
battery1.png (114.13 KiB) Viewed 1366 times
battery2.png
Sample using battery powered pre-amp: 300us pulse length (150us high and 150us low)
battery2.png (65.08 KiB) Viewed 1366 times
There’s not a great difference. The silences between the pulses are at the zero amplitude level indicating that some noise has been removed. However there does still seem to be a extremely consistent secondary double crest/trough characteristic throughout. Is this noise or is there any basis of this being a normal ZX81 characteristic?

User avatar
PokeMon
Posts: 2225
Joined: Sat Sep 17, 2011 6:48 pm

Re: Very low signal on MIC port

Post by PokeMon » Thu Jan 02, 2014 11:28 am

I think it is just the normal look of the signal.

Post Reply