Yep, the regulator is too hot to touch. I guess it wouldn't hurt to change it as well as the capacitors in the clock circuit.
- VCF South West - June 14 - 16, Davidson-Gundy Alumni Center at University of Texas at Dallas
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Southern CA
- VCF East - April 2025, Infoage Museum, Wall NJ
-
Please review our updated Terms and Rules here
Heathkit H89A Question
- Thread starter JimboX
- Start date
nothing to add, but following/subscribing. I picked up an H89A with the external floppy unit a couple of weeks ago at VCFSE. will be a while before I get to it, but it is such an awesome looking machine. watching this with interest.
daver2
10k Member
Have you got sensible readings first?
If you start randomly replacing things you can end up creating more problems than you solve if you are not careful. For example, you can damage the PCB.
If the regulator is too hot, this could be as a result of too high an input voltage - or the heatsink compound could have degraded to a point where it is not working reliably to dissipate the heat to the heatsink and the regulator is just overheating.
You are also quite correct in that it could be faulty, or there is too much current being drawn by the logic...
The only scenario here that would require the regulator to be replaced is if it was actually faulty.
Dave
You're right, Dave. It just gets frustrating at times. The capacitors in the Clock circuit are 2 Mylars and one that looks like a small diode C502 (.01 MFD). I Haven't seen this kind of capacitor before. These diode looking capacitors also filter the input to the 5-volt regulators. They're electrolytics of 1MFD. I don't know if I should change any of them or not. C505 is also this weird looking capacitor(.01MFD). When I measure the filter caps at the 5- volt regulator I get 47 ohms to ground. I'm not to sure what to do at this point. Should I still look for noise at the 5-volt regulator? Thanks.
daver2
10k Member
>>> Should I still look for noise at the 5-volt regulator?
Yes, and around the ICs.
If the capacitor you are measuring is across the +5V and 0V rails, you are NOT measuring the resistance of the capacitor, but everything else hung off the power rails!
It is extremely unlikely for these low-valued capacitors to actually fail - so a red herring. Tantalum bead and electrolytic capacitors are your "go to" capacitors for this.
Dave
Yes, and around the ICs.
If the capacitor you are measuring is across the +5V and 0V rails, you are NOT measuring the resistance of the capacitor, but everything else hung off the power rails!
It is extremely unlikely for these low-valued capacitors to actually fail - so a red herring. Tantalum bead and electrolytic capacitors are your "go to" capacitors for this.
Dave
Okay, I'll keep checking for noise. Is there a name for those diode looking capacitors? Just wondering.
I've been measuring the noise levels, but it's hard to do so when the signal moves around so much. Is there some way to stabilize the signal? From what I can get, the noise seems to be in the low millivolt range. While checking, I found that the chip U502 has an intermittent signal out on pin 8. The signal coming from U501 Pin 4 is strong and constant and its signal to U502 pin 1 is also strong and constant. Should I assume this chip is faulty since the signal to pin 8 originates from within the chip itself?
Did you give up on me, Dave? I wouldn't blame you if you did. I solved the voltage problem by removing the plug-in cards. The 5v is study now. I'll worry about those cards later. The interrupts on the CPU board are now reading 5 volts. The interrupts on the terminal logic board are still reading 0 volts. I found a few bad chips that need replacing. Thanks for your help, Dave in steering me in the right direction.
daver2
10k Member
No I haven't given up on you. Just been rather busy with work. It is a public holiday in the UK, and I am trying to catch up with a load of outstanding work.
What are the states of the /M1, /MREQ, /IORQ, /RD and /WR CPU pins of both the main and keyboard Z80 processors now? HIGH, LOW or PULSE activity.
The 'stuck' interrupt line on the keyboard Z80 processor could come from a number of sources. Check U447 pins 4, 5 and 6 and U432 pins 8, 9 and 10 for HIGH, LOW or PULSE activity.
Dave
What are the states of the /M1, /MREQ, /IORQ, /RD and /WR CPU pins of both the main and keyboard Z80 processors now? HIGH, LOW or PULSE activity.
The 'stuck' interrupt line on the keyboard Z80 processor could come from a number of sources. Check U447 pins 4, 5 and 6 and U432 pins 8, 9 and 10 for HIGH, LOW or PULSE activity.
Dave
Last edited:
There's no hurry on anything, Dave. I just appreciate your advice when you're able to give it. I'll get these readings tomorrow. I did find chip U446 section F on the keyboard encoder was bad. When the control key was pressed, the 5 volts from pin 13 wasn't transferred to pin 12and thus to pin 19 on the encoder. All the other sections on this chip worked properly when the reset and shift keys were pressed. I'm ordering some new chips to replace it and a few other suspected chips.
Here's the pins from U430. M1 not used. MREQ 0 volts. IORQ pulse. RD 0volts. WR 3.86 volts. Wait Pulse. Clock Pulse. Interrupt 0volts. U504 Clock Pulse. Interrupt 4.48 volts. NMI 5.02 volts. MREQ Pulse. IORQ 3.82 Volts Pulse. RD 2.26 volts with Pulse. WR 3.82 with pulse. M1 2.97 with pulse. Reset 3.0 volts Pulse. U447 Pin 4 0 volts Pin 5 0 volts . Pin 6 5 volts. U432 Pin 8 0 volts. Pin 9 4.5 volts. Pin 10 0 volts. No voltage to RP4(2) JP12 pin 1 and 2. feeds pin 4 on U447. Continuity to U449 pin 3 and U445 pin 17 good. Hope this helps. Thanks.
daver2
10k Member
Correct, even though /M1 is not connected, it is an OUTPUT from the Z80 and is still used to indicate that the CPU is fetching (and executing) instructions.
/MREQ being 0V is not correct either.
/RD being 0V is not correct either.
Again, how are you reading these signals? Some of the voltages you are getting are really weird. They are supposed to be TTL levels...
I'll check the other pins shortly...
Dave
/MREQ being 0V is not correct either.
/RD being 0V is not correct either.
Again, how are you reading these signals? Some of the voltages you are getting are really weird. They are supposed to be TTL levels...
I'll check the other pins shortly...
Dave
Last edited:
daver2
10k Member
U504 (Z80) /RESET pin pulsing is not good.
You have a standing /INT on U430 (Z80) with this pin being LOW (0V).
U447 pins 4 and 5 being both LOW indicates two (2) interrupting sources. Pin 4 is from the keyboard encoder and pin 5 is from the <BREAK> key.
Pin 5 SHOULD be HIGH if the break key is not depressed. This should be simple to chase back and find out why this pin is stuck LOW.
Dave
You have a standing /INT on U430 (Z80) with this pin being LOW (0V).
U447 pins 4 and 5 being both LOW indicates two (2) interrupting sources. Pin 4 is from the keyboard encoder and pin 5 is from the <BREAK> key.
Pin 5 SHOULD be HIGH if the break key is not depressed. This should be simple to chase back and find out why this pin is stuck LOW.
Dave
U430 is the microprocess on the terminal logic board. I'll check out M1 to see if it's pulsing. I'll also check out the other pins to see if they're pulsing or I'm reading something wrong. Where I read zero on some of the pins, they are in the microvolt range so I am calling them 0. Maybe that's wrong. I'll check out U447 pin 4 and 5 again.
daver2
10k Member
As I have stated myself previously - if you have an oscilloscope (and I think you do) then only use this. A logic probe is not guaranteed to work at these frequencies and a multimeter voltage reading is as good as useless.
A logic '0' is defined to be less than 0.5 Volts and a logic '1' is defined to be greater than 4.5 Volts. These are my 'sensible' thresholds and not the official TTL levels.
Anything between these two voltage levels (as indicated on the oscilloscope screen) would be considered suspect.
Dave
A logic '0' is defined to be less than 0.5 Volts and a logic '1' is defined to be greater than 4.5 Volts. These are my 'sensible' thresholds and not the official TTL levels.
Anything between these two voltage levels (as indicated on the oscilloscope screen) would be considered suspect.
Dave
I remeasured the pins on the terminal logic board microprocessor U430. Pin 19 MREQ has a pulse. Pin
21 RD Has a pulse. M1 has no pulse but reads 3.88volts. The interrupt has reads in millivolts. IORQ has a pulse. U447 Has 5 volts on pin 5 and 6. Pin 4 reads .333 millivolts. Not sure if it's supposed to read this. Thanks.
21 RD Has a pulse. M1 has no pulse but reads 3.88volts. The interrupt has reads in millivolts. IORQ has a pulse. U447 Has 5 volts on pin 5 and 6. Pin 4 reads .333 millivolts. Not sure if it's supposed to read this. Thanks.