exxos's Atari, Amiga & retro forum

My stock falcons floppy drive malfunctions the longer the machine is turned on. I put that down to missing pull ups as well.

@exxos Seems like pull ups are responsible for a lot of what ails the ST, I certainly replaced them all on my Megas recently, and this board will be getting them, but time is my enemy at the moment as work is very busy, I have to visit my elderly parents this weekend and half of next week, then the wife insists I spend time with her!

I'll get there in the end.

Now I'm working from home I'm managing to fit odd quick bits in here and there. LS06 changed over lunch, no different floppy still gets a busy light and no desktop, so DMA an AY next on the list. If there is no difference after that then time for those SILs to bet bounced for new and add in the DMA/Floppy pullups.

I hate it when stuff just stops working.

rubber_jonnie wrote: 12 Mar 2020 13:29 Now I'm working from home I'm managing to fit odd quick bits in here and there. LS06 changed over lunch, no different floppy still gets a busy light and no desktop, so DMA an AY next on the list. If there is no difference after that then time for those SILs to bet bounced for new and add in the DMA/Floppy pullups.

I hate it when stuff just stops working.

Same here, happens all the time :( but hopefully those days are over now the H4 is born :) I think there are just so many problems with the original boards it's a wonder they ever worked at all.. I think the soldering is just simply bad, but when you start manhandling the boards a bit, the soldering gives up in multiple places on that board never works again. I think the problems with the original ST's is simply " in its bones". :(

2 of my H4's are sat on the shelf.. They was fully working when they was put on the shelf, so they are currently going to sit there for a couple of weeks, and the test will be if they still work when they come back off the shelf...

I'm not sure this qualifies as "soak testing" as they are actually powered off, but that is actually what is happening :lol: "un-powered soak test" maybe :shrug:

exxos wrote: 12 Mar 2020 13:39 Same here, happens all the time :( but hopefully those days are over now the H4 is born :) I think there are just so many problems with the original boards it's a wonder they ever worked at all.. I think the soldering is just simply bad, but when you start manhandling the boards a bit, the soldering gives up in multiple places on that board never works again. I think the problems with the original ST's is simply " in its bones". :(

When you use the scope on the H4 its pretty obvious that the board is much cleaner signal wise than the original Atari efforts. My mega's signals are exceptionally noisy compared to the H4's. These board are a big step forward. :thanksyellow:

PaulJ wrote: 12 Mar 2020 14:02 When you use the scope on the H4 its pretty obvious that the board is much cleaner signal wise than the original Atari efforts. My mega's signals are exceptionally noisy compared to the H4's. These board are a big step forward. :thanksyellow:

I have just been scoping the bus out, and it is indeed a lot cleaner than the original boards which look like they are powered by LSD mostly.

exxos wrote: 12 Mar 2020 13:39
Same here, happens all the time :( but hopefully those days are over now the H4 is born :) I think there are just so many problems with the original boards it's a wonder they ever worked at all.. I think the soldering is just simply bad, but when you start manhandling the boards a bit, the soldering gives up in multiple places on that board never works again. I think the problems with the original ST's is simply " in its bones". :(

2 of my H4's are sat on the shelf.. They was fully working when they was put on the shelf, so they are currently going to sit there for a couple of weeks, and the test will be if they still work when they come back off the shelf...

I'm not sure this qualifies as "soak testing" as they are actually powered off, but that is actually what is happening :lol: "un-powered soak test" maybe :shrug:

@exxos lol I'm not convinced that is soak testing either!!

I'm really pleased that my H4 is good after never really arriving with the Alpha board. It's also useful as I can test any chip in the H4 as they're all socketed.

I think I do need to get another H4 though after all the failures along the way for various reasons. I had to put a bodge wire in to the cart port because a short burned out one of the tracks, and then there's the LS11 bodge wire because the soldeing didn't stick to the pad for whatever reason. I'd like a nice unblemished one :)

As for manhandling, I think that is an underestimated source of errors/faults, especially given the quality of the PCBs and soldering. I have seen a lot of boards with untrimmed component leads that can get bent over too, though thankfully never had one cause a problem. Yet.

Problem I have now is running out of parts. I've had a couple of dead GLUEs lately, and I have no spare Blitters at all, and we all know how much they cost!! Hopefully I'll win a bid on a knackered old ST tonight and either that or my faulty test board can become a donor.

exxos wrote: 12 Mar 2020 14:16

PaulJ wrote: 12 Mar 2020 14:02 When you use the scope on the H4 its pretty obvious that the board is much cleaner signal wise than the original Atari efforts. My mega's signals are exceptionally noisy compared to the H4's. These board are a big step forward. :thanksyellow:

I have just been scoping the bus out, and it is indeed a lot cleaner than the original boards which look like they are powered by LSD mostly.

Yeah baby...

:girldance: :girldance: :girldance:

rubber_jonnie wrote: 12 Mar 2020 14:24 I think I do need to get another H4 though after all the failures along the way for various reasons. I had to put a bodge wire in to the cart port because a short burned out one of the tracks,

Oh what track was that ? I did half wonder the other day if I ought to add a fuse into the power to the cartridge port..

rubber_jonnie wrote: 12 Mar 2020 14:24 As for manhandling, I think that is an underestimated source of errors/faults, especially given the quality of the PCBs and soldering. I have seen a lot of boards with untrimmed component leads that can get bent over too, though thankfully never had one cause a problem. Yet.

Yeah I have seen things not could probably and bending over shorting out other things as well. I think I documented in my blog a few weeks back that I had to re-solder literally half of a motherboard because the soldering was that bad. The machine was otherwise untouched virgin... So I think these machines are just fail for no reason is likely because they have had a bit of the bash about and is broken the solder, I think it's just pretty conclusive these days. There simply is not going to be many of the original boards working at all in the future..

rubber_jonnie wrote: 12 Mar 2020 14:24 Problem I have now is running out of parts. I've had a couple of dead GLUEs lately, and I have no spare Blitters at all, and we all know how much they cost!! Hopefully I'll win a bid on a knackered old ST tonight and either that or my faulty test board can become a donor.

Yeah, when I list spare GLUE's etc in my store someone snaps them up normally pretty quickly. I can only get them from basically scrap machines which are costing like £70+ these days. Getting money back for basically machines just used as spares is pretty tough going...blitters are impossible to find :( The price for them is very high because that is pretty much the price I had to pay to buy them in the first place :(

Again it is the reason @Icky is busy developing the FPGA boards... I mean because the MMU,GLUE,BLITTER is getting on for like £100 these days, a FPGA board I think can be built for half the price... Problem is currently the cause or not 100% compatible. But this may be fixable one day.

@exxos Yeah, you were luck to find those Blitters, don't know how you did it, but I'd like at least 2 or more and I have other stuff that would be more useful. They're very much a nice to have. My H4 doesn't even have one! Had to borrow one out of a Mega to test it.

Prices of spares or repairs machines are just insane right now, gone are the days it seems when you could get a board for £10 off ebay. I'm annoyed too, I had a dead one, PCB only, which I binned when we moved house 6 years ago and I'm pretty sure I binned it complete, chips and all.

The damn thing powered up to a white screen, so I think given my current experience, I could probably fix it now. Not to be. And as you say, they will be slowly dying off as time goes by. My oldest machine, a big button STF, has a PSU in need of recapping, so I put in one that was already done. Then I found if I tightened up one of the screws that passed through the leg of the PSU up too tight, the machine won't power on! Go figure.

Fingers crossed @Icky Can move forward with the FPGA stuff, it would be good if we could have an all in one FPGA GLUE, MMU and Blitter replacement, or even three separate devices I'd be happy.