Previous Thread
Next Thread
Print Thread
Page 2 of 3 1 2 3
Re: Bondwell BW-14 [Re: rfka01] #116492 11/26/19 02:39 AM
Joined: Jan 2012
Posts: 814
C
crazyc Online Content
Senior Member
Online Content
Senior Member
C
Joined: Jan 2012
Posts: 814
After cpm boots it's disk reading routine has a long delay look before it checks for completion. That loop is long enough to take several revolutions to complete. It seems like there ought to be an interrupt to break out of the loop but the way the fdc irq is connected don't appear to work like that.

Re: Bondwell BW-14 [Re: rfka01] #116493 11/26/19 07:03 AM
Joined: Jan 2012
Posts: 1,028
rfka01 Offline OP
Very Senior Member
OP Offline
Very Senior Member
Joined: Jan 2012
Posts: 1,028
The schematics I posted to the FTP came from here. The page about the FDC contains a FDCINT line.


NCR DMV- DEC Rainbow- Siemens PCD- ITT 3030-Oly People- Acorn A5000- Olivetti M20
Re: Bondwell BW-14 [Re: rfka01] #116494 11/26/19 12:42 PM
Joined: Mar 2001
Posts: 16,391
R
R. Belmont Online Content
Very Senior Member
Online Content
Very Senior Member
R
Joined: Mar 2001
Posts: 16,391
Right, FDCINT goes to PA7 on the 6821.

Re: Bondwell BW-14 [Re: rfka01] #116495 11/26/19 04:07 PM
Joined: Jan 2012
Posts: 814
C
crazyc Online Content
Senior Member
Online Content
Senior Member
C
Joined: Jan 2012
Posts: 814
Page 1 shows a 556 timer connected to the motor on line. We are turning off the motor immediately which causes each read to do turn it back on and wait for a bit until it spins up. The timer is supposed to hold the motor on for bit which allows it to skip the loop. The simple fix is to just never shut off the motor, better is to figure out approximately how long the timeout should be.

Re: Bondwell BW-14 [Re: rfka01] #116496 11/26/19 05:15 PM
Joined: Mar 2001
Posts: 16,391
R
R. Belmont Online Content
Very Senior Member
Online Content
Very Senior Member
R
Joined: Mar 2001
Posts: 16,391
The schematic doesn't exactly match any of the usual 555/556 hookup styles, so it may be doing something weird. Maybe mock it up with the netlist system and see what happens?

Re: Bondwell BW-14 [Re: rfka01] #116497 11/26/19 07:19 PM
Joined: Mar 2006
Posts: 1,035
L
Lord Nightmare Offline
Very Senior Member
Offline
Very Senior Member
L
Joined: Mar 2006
Posts: 1,035
Looking at the 556 circuit at http://www.eld.leidenuniv.nl/~moene/Home/museum/Bondwell12/schematics/BW12-A1-CPU-IO-Decode.pdf, I'm trying to figure out what they were trying to do here.

It seems they basically used the 556 as a fancy voltage threshold comparator with hysteresis; the IC51 7416 open collector inverter pulls the TRIG and THRESHOLD pins low when its input (MOTOR1 | MOTOR2) is high, which makes the OUT pin of the 556 high and leaves it that way (which turns on the motor). When the (MOTOR1 | MOTOR2) signal is low, the 7416 allows the TRIG/THRESHOLD pins to be charged (via the 100k resistor at R16 to VCC, charging the capacitor at C11 which is, I think, 1.7uf electrolytic) until the THRESHOLD pin reaches 2/3 of VCC at which point the 556 shuts off and turns off the motor.

This is a non-standard hookup of the 556, since normally the capacitor is between the threshold/trigger pins and ground, not between them and VCC. The 556 is basically being used only for hysteresis (turns on when trigger is below 33% of VCC, turns off when threshold is above 66% of VCC) with the trigger and threshold pins tied together, and the external circuit with the 7416 discharging C11 and C11 recharging through the 100k resistor.

Since the 7416 pulls directly to ground it should discharge C11 very fast (and this current surge is probably not great for the lifespan of either C11 nor the 7416, though the 7416 is meant for very high voltage applications so it may be ok), however this discharge rate technically controls how fast the motor will turn on after request, so it is a finite (but very small) delay, which might be able to be ignored.

So, for the old fashioned τ = RC time constant equation for how long a capacitor takes to charge to ~63% of its capacity (and assuming 63% and the 66% of vcc threshold for the 556 are 'close enough' to be the same), the motor off delay is ~0.17 seconds.

This should be easily emulatable using a timer.


"When life gives you zombies... *CHA-CHIK!* ...you make zombie-ade!"
Re: Bondwell BW-14 [Re: rfka01] #116498 11/26/19 08:12 PM
Joined: Jan 2012
Posts: 814
C
crazyc Online Content
Senior Member
Online Content
Senior Member
C
Joined: Jan 2012
Posts: 814
170ms appears to be enough to keep the motor on between consecutive sector reads.

Re: Bondwell BW-14 [Re: rfka01] #116499 11/26/19 08:30 PM
Joined: Mar 2006
Posts: 1,035
L
Lord Nightmare Offline
Very Senior Member
Offline
Very Senior Member
L
Joined: Mar 2006
Posts: 1,035
A possible modification that this circuit could have to reduce wear on C11 is adding a resistor between the 7416 output and the resistor/capacitor/threshold/trigger pins, to somewhat reduce the amount of current the capacitor is dumping out through the 7416 during discharge; this would have to be carefully calculated so that the motor on timing doesn't get delayed by so long that it starts causing read errors.


"When life gives you zombies... *CHA-CHIK!* ...you make zombie-ade!"
Re: Bondwell BW-14 [Re: rfka01] #116500 11/26/19 08:38 PM
Joined: Jan 2012
Posts: 814
C
crazyc Online Content
Senior Member
Online Content
Senior Member
C
Joined: Jan 2012
Posts: 814
It's already got quite a long delay loop waiting for the motor to start. A few extra ms for the cap to discharge probably wouldn't cause any problems.

Re: Bondwell BW-14 [Re: rfka01] #116501 11/26/19 10:37 PM
Joined: Jan 2012
Posts: 1,028
rfka01 Offline OP
Very Senior Member
OP Offline
Very Senior Member
Joined: Jan 2012
Posts: 1,028
Thanks to all involved!
Booting and formatting work like a charm now. Copying the system (COPYSYS) and PIPing files over from one disk to another still fail. I'll try to find other systems that use a 18 sectors/track, 256 bytes/sector format and check those commands there.

Last edited by rfka01; 11/26/19 10:41 PM.

NCR DMV- DEC Rainbow- Siemens PCD- ITT 3030-Oly People- Acorn A5000- Olivetti M20
Page 2 of 3 1 2 3

Who's Online Now
1 registered members (reenigne), 65 guests, and 2 spiders.
Key: Admin, Global Mod, Mod
ShoutChat Box
Comment Guidelines: Do post respectful and insightful comments. Don't flame, hate, spam.
Forum Statistics
Forums9
Topics8,722
Posts114,638
Members4,873
Most Online510
Aug 26th, 2019
Powered by UBB.threads™ PHP Forum Software 7.7.3