Spectrum 128k Toastrack. Investigation and modification of this British computing icon.
Вставка
- Опубліковано 9 лип 2024
- This week we are looking at a Spectrum 128k toastrack that belongs to @YawningAngelRetro. He asked if I could sort out a power issue and carry out a few mods so lets see what we can do.
The mods completed in this video are a composite mod with audio and colour bleeding fix. Audio level fix, a jail bar fix and we also fit a new regulator.
All components were purchased from www.retroleum.co.uk/
High quality spectrum schematics can be found at this link www.8bit-wiki.de/fileadmin/8b...
UA-cam channels mentioned in this video:
Yawning Angel (Amiga AMOS tutorials) - / @yawningangelretro
Rose Tinted Spectrum (Vintage gaming shenanigans ) - / @rosetintedspectrum
CRG Links
CRG X (Twitter) - / yt_crg
CRG Patreon - www.patreon.com/user?u=13919007
CRG Discord - / discord
Chapters
00:00 Intro
00:30 Letter
02:38 Opening up
04:47 Composite mod
15:03 Components
16:20 Colour bleeding mod
18:33 Jail bar mod
21:27 Audio level mod
22:29 Replacement regulator
26:43 Revisiting the jail bar mod
32:40 Testing and printer problems
34:34 Tape tester
36:35 Dizzy - Наука та технологія
Of all the retro channels on UA-cam, this is my favorite. I love how there's no "big theme", it's just a normal man in a a basement fixing some old computers. I love it.
Thank you, it means a lot to hear people enjoy the channel.
Glen, different types of capacitors suit different frequency bands. It's a vast topic and I would be happy to go through it with you but for this comment let's just say
- no capacitor is ideal, they all have parasitic properties (resistance, inductance)
- parasitic inductance makes their behaviour transition from 'capacitive' (= impedance decreasing with frequency) to 'inductive' (= impedance increasing with frequency) and there's a frequency which marks this change of behaviour
- if you take different capacitors with the same nominal capacitance but different parasitic inductance you'll get different frequencies at which the behaviour changes. The parasitic inductance is linked to the physical structure of the component so wider/larger is associated with lower inductance, longer terminals/legs (through-hole vs SMD) with higher inductance.
- placing components in parallel gives you the sum of the capacitance and the 1/Lp = 1/L1 + 1/L2 of the inductance (kind of like for resistors)
- as technologies, electrolytic generally offer the highest capacitance for component but ceramic offer the lowest parasitic inductance per nominal capacitance. This means that at "high" frequencies electrolytic cap are likely to have a significant higher inductance than at lower frequencies while ceramic ones keep decreasing (so to speak). At 10s of MHz the nominal capacitance value of electrolytic does not give you a good guide of their actual behaviour while ceramic capacitors are typically characterised for such frequencies and, even if their nominal capacitance is lower, the effective impedance is lower than the one of electrolytic.
- knowing the frequency band at which you want the component to show low impedance can guide you in the choice of technology and different capacitors in parallel work together to give you the desired impedance in each band of interest
I hope not to have confused you!
Totally confused but fantastic information, thank you. I do understand some of it but a lot of it goes way over my head. I'd need to go back to school to learn all this stuff properly.
@@CRG
I'm sure that, step by step, you can learn by trying new things. By the way I really like how you're able to sort out the many troubles retro stuff throws at you.
My comment was inspired by the mode at 27:26 when each 22nF was replaced by 1uF. No change of dielectric material (X7R, I suppose) so the effects of the increase on frequency response should be pretty straightforward as the parasitic inductance has similar order of magnitude. But if you had to go for lower voltage rating then the smaller package would come with higher parasitic inductance than the larger voltage/package.
The traces between these decoupling capacitors and the ICs goes kind of roundabout way and that doesn't help with higher frequencies.
Purely out of curiosity, I would like to see what the effect would be of fitting a pair of smaller capacitors (let's say 2x 220nF or 2x 470nF) in parallel directly between the VCC-VSS pins of each of the memory IC (the pinout is unfortunate because the distance between the pins is as large as it can be in a DIP-16). It would require a lot of work and it would not look pretty, I admit.
Hi Glen, good to see this , as I just bought a 48k Plus as spares/repair, and re-capped it and got it working. Then I done the 5v regulator mod, and the jailbar mod also. I didn't find the jailbar mod made a lot of difference though. Also the composite mod, but used a capacitor across the output. Very interested to see if the 1uf axial caps would improve my one too... I'll drop it round to you :)
I don’t think we were hearing the AY chip there. I’m not 100% sure but I think the AY chip was only really used on 128k games. The AY has a much “fuller” sound than what we heard there. Pretty impressive of the coders to get what sounded like multi voice audio out of the old beeper. Some of the later Dizzy games used the AY chip, like Fantasy World Dizzy for example.
your correct. as far as i know the ay chip is only enabled in 128k mode and for example the music in the likes of fantasy world dizzy loaded extra content into other banks of memory that were not available on the 48k model. you could enable the ay chip in 48k mode on a 128 by going into 128 basic and issuing the command randomize usr 0 and the machine will reboot into 48k mode which leaves the memory paging and the ay chip accessible
Just goes to show I know very little about these machines. Ah well, at least I know that now.
Great work! Only thing I might have done differently is to replace the regulator with a new 78S05 rather than a switch mode regulator. That Traco may be the reason you were seeing 10-30khz noise with the oscope, and could potentially contribute to the jailbars.
I may try that as I'll have the machine here for another week or so yet. Thanks for the tip.
The music on Dizzy is actually coming from the ULA/beeper... Yes it is possible to emulate multi-channel sound on the 48 (kind of!)
Yeah, seems I was mistaken but at least I know now.
Adding that many 1uF caps could cause the regulator to take longer to start up. a modern switcher with "soft start" would help.
The Traco is a switched regulator although I'm not sure on its starting method.
@@CRG most switching regulators have some "soft start", but special ones allow you to add a capacitor to a IC pin to slow down how fast they ramp up the voltage. If the voltage is ramped up too fast for a given amount of capacitance the regulator will go into current limiting and "hickup mode" this will cause it to take longer to turn on. The only problem would be making sure the reset signal is delayed enough.
good video..glenn ...i thoght you did a good job!..those jailbars seem to be somewhat of a pain to eliminate!
Thanks Mike. Not gotten rid of the jailbars completely but they certainly are a little better than when I started.
@@CRG absolutely
The 128 +2 has the voltage regulator on a heatsink and connected to the board with a plug and wires like the toastrack.
The +2 is essentially just a repackaged toast rack, the circuit is virtually identical and even the ULA is compatible. The +2A/+3 is a very different beast.
Thanks for the info, these machines are all new to me. I didn't know the +2 was just more or less the same thing.
Those little rotary resistors are notorious for breaking once they get old. Trash 80/Dragon computer use the same parts and they break all the time.
Nice.
I made the sound mod to my toast rack now I don't get any sound at all :(
Only thing I could suggest is you maybe fitted the wrong resistor or accidently bridged something?
@@CRG I put the original back in and it still didn't work 😭. Will try again at the end of the week.
Thumbs up from me - nothing Amiga related
I do want to get back to a variety of stuff, things were very heavily Amiga focused for a while. That's not to say there won't be Amiga content but I do want to spread it out a bit more.
The music in the game was not coming from the AY chip. You switched to 48K mode which does not access the AY chip. That was all pure Spectrum 48K buzzer. You can also very easily tell when the music is AY as it's a lot clearer.
Mistake on my part and thanks for pointing it out.
The audio cable holding you did; was that shorting both channels to make a mono type mode?
A mono jack shorts the right channel to ground... Pulling it out slightly moves the whole thing up one band so the right channel takes the place of what would have been the left, and the ground is isolated from the middle ring.
Yep, just as Jason said above it was just shorting the channel to make connection through to the computer.
That Traco regulator is 1A rated, it is a bit borderline for the Toastrack, better to use the 1.5A version, or even better, 2A unit from Recom.
Do you mind me asking how you know it's 1A rated? As I said this was ordered from retroleum and was listed as a 2A part there so if indeed 1A I'll have to take that up with them.
@@CRG I may be mistaken if Retroleum sells them as 2A devices, but AFAIK, Traco does not have a 2A 5V regulator in this package. Can you please search with its part number and check the data sheet?
@@borayurt66 I have a Traco TSR-2-2450 in my Sinclair QL, that's 2A rated, according to all specs that I can find, and it works fine. That's also what Glen uses here. It's fine.
@@chainq68k OK, as I told Glen, I am not aware of a Traco device in that package which is 2A rated. Maybe it is a new product, I know about a 1.5A rated Traco, which is almost double the size lengthwise. Unfortunately I can't make out the part number from the video, it was the size that got me worried.
@@borayurt66 @25:54 if you set the video to full HD, and zoom in, you can clearly see it's a Traco TSR-2-2450.
Nice job with spectrum image quality finetuning. But, is it just me, or is there a slight postprocessing blur forgotten througout most of the video? E. g. 3:33 around G, V, B Keys.
I noticed that myself, I'm not sure what caused it but it's like the centre of the image is a little blurry. I'll keep an eye on it to be sure it's not a fault with my camera. I did check the lens and it's clear so I'm not sure what it is.