1050 Happy Problems

[TGB1718]

I've got an issue with a 1050 Happy Drive in as much as it will sometimes work fine, but the it sometimes fails to boot

and will sit there with the disk spinning requiring a switch off/on to reset it.

 

The problem I have is that after powering back up, the drive doesn't do the usual head movement to track 0 (with or without a disk inserted)

and it can't be seen by the 800 or 130XE, the strange thing is if I leave the drive OFF for some time > 10 minutes the drive will come back and

will boot, sometimes it will survive a turn off/on but it will go back into that state eventually.

 

I have tried removing the power lead, leave a minute or so and while disconnected, turn the drive on to discharge any Cap's, turn it back on, but

it will still be unresponsive.

 

I think the issue with boot problems is due to missalignment, I will be doing a realignment, but wanted to stop the lack of response when turning off/on.

 

I monitored the +5V and +12V both are well within limits, however when I trun the drive off, the 5V goes immediately but the 12V discharges very slowly

and stays around 0.7V for ages. I shorted the 12V line with the power off, removed the short, power the drive back on and it started normally.

 

Anyone know if this is a "feature" of Happy Drives, I have included a photo of the board in my drive. 

 

Was thinking of putting a 600 1/4W ohm resistor across C65 to help discharge the residual voltage

 

I replaced all the Cap's some years ago.

[+Nezgar]

My first suspicion is flakey/bad SRAM, so try replacing that 8164A the chip (8Kx8). 6264 should also be compatible.

 

There's a chance it could also be a flakey 6810 SRAM chip on the motherboard (U8 socket).

 

Also, it could be flakey connection between the socket on the motherboard and the happy board. The original sockets are poor quality single wipe, and your happy board appears to use round machine pins, so not much contact area between the socket wipe and the pins. To rule out that completely, you could replace the CPU socket (U9) with a machine socket.

 

Also, just to be sure have you run the diagnostics tests in the Happy 7.1 utilities to start with? Do you ever have problems once it starts OK, or only the power off/on symptom.

 

I'd also test further by removing the happy board and replacing a stock 6507 & ROM and verify stable operation that way as well. Then you can also run Atari's 1050 diagnostics software.

[_The Doctor__]

Most of the time a wiggle of the board in the socket clears this issue on old drives for me, sometime taking them out and re insertion can damage the old frail sockets.

remember to leave sufficient time between power down and up to let all of the energy dissipate for a clean power up.

same things go for some computers as well

[TGB1718]

I had removed the board and re-inserted it, lifted and re-inserted IC's, I didn't receive the original ROM when I bought this,

but I think I remember seeing a chip with TANDON printed on it which I think came from my own 1050 when I put a US Doubler

in it, time to hunt for that chip and convert it back to normal 1050.

[TGB1718]

On 4/29/2020 at 4:49 PM, Nezgar said:

My first suspicion is flakey/bad SRAM, so try replacing that 8164A the chip (8Kx8). 6264 should also be compatible.

 

There's a chance it could also be a flakey 6810 SRAM chip on the motherboard (U8 socket).

I had a closer look at the RAM chip to see if I could test it, it's an 8464A which has the same pinout as the 2764 EPROM.

 

Recently I built an EPROM programmer running off a Raspberry Pi and used it to burn some games onto cartridges.

It can burn 2732, 2764, 27128 and 27256 EPROM's

 

As I wrote the software to run it, it shouldn't be too hard to add a procedure to test this RAM chip.

I'm also building a little adaptor so I can test the 6810 RAM chip too as the pinouts are very different on this chip

as I had to for the 2732.

 

The only difference on the programmer is that pin 27 on the EPROM is address A13  and on the RAM its CS2, so just

a matter of setting the start address to 8192 to enable CS2 and clock from there

[blacka013]

You should also check the value of the resistor, as it is not 4k7 ohms.

[TGB1718]

27 minutes ago, blacka013 said:

You should also check the value of the resistor, as it is not 4k7 ohms.

Nice spot, I'll check where it goes, (have to open the drive up again), it does look like a 26K thats

being used as a pulldown as one end is going to the GND pin of the RAM chip.

[TGB1718]

Had a look at the board, that resistor is a pull-down on the output of a 74HCT10 which is under the RAM chip, this output

is also connected to pin 20 !CS1 , it's possible that this resistor is too high a value, maybe change it for 4k7 and see if things

improve, certainly won't break anything.

[blacka013]

It certainly can't hurt. I have this version of the Happy board and I found some of the soldering left a lot to be desired.

 

Does your board have a resistor pack across pins 8 - 16 of the 6502?

[TGB1718]

1 minute ago, blacka013 said:

Does your board have a resistor pack across pins 8 - 16 of the 6502?

No, that resistor was the only one on the board, I did replace it with a 4k7 resistor, made no difference.

 

Will wait now for a 6507 that I ordered so I can turn the drive back to a standard 1050, then see how it goes.

[+Nezgar]

1 hour ago, TGB1718 said:

Will wait now for a 6507 that I ordered so I can turn the drive back to a standard 1050, then see how it goes.

In the mean time while you wait for a 6507, you can try this trick: Remove the ROM from the happy board, and install the original Tandon/1050 ROM on the 1050's motherboard. It should now function as a stock 1050, using the 6502 CPU on the happy board, and the ROM & RAM on the 1050 motherboard. The stock 1050 OS will only use the 6810 RAM, and not the Happy's RAM even though it's still technically present. (it should still work with that removed too though heh). The CPU's are compatible, the 6507 just has a more limited address space.

 

I am reminded of this as one of my 1050's had a clever mod installed: Both ROM's were present, but Pin 24 (Vcc) was lifted out of each, wired to an SPDT switch (2 positions, 3 terminals), which chose which of the two ROMs was active receiving Vcc... Effective true "unhappy" switch - obviously best switched with the drive OFF.

[TGB1718]

 

42 minutes ago, Nezgar said:

In the mean time while you wait for a 6507, you can try this trick:

Nice trick, sort of obvious when you think about it, will have a go tomorrow

[TGB1718]

Ok, did the business, powered on, nothing, just the power on led.

 

Had a look and found the jumper settings were set for PROM, my TANDON chip is an EPROM, so had to strip it to the circuit board,

swapped the jumpers over, put back together, powered up with the 1050 diag disk in and it booted straight in, no problems.

 

Ran the diagnostics check, passed all tests except for Track 0 sensor which I expected as I replaced the sensor some time ago and

ran the alignment setup when it had the Happy board in and I wasn't convinced the alignement was exactly right, good enough to

boot, but probably not good enough for all disks, so will run the alignement again.

 

Also ran the motor speed test and it's a little slow, so will fix that too.

 

If I can't find out whats wrong with the Happy board, I will probably not re-fit it, but will install a US Doubler, I already have a spare ROM

and have some 6810 RAM chips on order.

 

Many thanks @Nezgar for pointing out I could run this in "UnHappy" mode this way

[blacka013]

Hi TGB1718.

 

if you can't find the problem with the board, I'm just up the road south of Lincoln if you want me to have a look at it for you.

 

Regards

[TGB1718]

Managed to buy 5 6810 chips on Ebay, turned up today, tested them all, 4 out of 5 working, so tomorrow this happy

becomes a US Doubler and see how it goes

[TGB1718]

6507 CPU arrived today, did the "piggy back" job to the RAM chips, used my spare USDBL ROM,

drive started normally, formatted SS/DD using SDX and it's currently copying files from my SDrive-Max

to a "real" floppy running at USDBL high speed. (I do love the sound of real floppy drives doing their thing).

 

I also removed the RAM chip off the Happy board, it appears to be ok, but will give it a soak test, I thought this may have been

the problem as the 6502 ran the drive normally with the Atari Tandon ROM inserted, so eliminating the 6502 as a possible problem.

 

So maybe the ROM is faulty, I'll see if I can get one and program it, just because I would like to know what the problem was.

 

For now I'm really happy (excuse the pun) as I have 2 working USDBL drives, I only ever used the Happy for it's High Speed and DD, not for copying.