Atari 400 power on successfully only now and then

[Mclaneinc]

Well logic says there's an issue with the modding..A small triple check of the area may sort the issue out...Even the best of us make the odd dry joint or bridge somewhere, needless to say I have connected to the wrong point a few times..

[mimo]

48 minutes ago, Mclaneinc said:

Well logic says there's an issue with the modding..A small triple check of the area may sort the issue out...Even the best of us make the odd dry joint or bridge somewhere, needless to say I have connected to the wrong point a few times..

I agree, until the original issue is resolved all you will do is damage a new power board 

 

[slaanesh]

I’ve double and triple checked the mods and am sure they are good with no spillage, bridges or misconnections.

[slaanesh]

So I have a new power board.

 

Before using it I checked my power supply to make sure that it was providing a good constant 9VAC. It's actually providing bang on 10VAC which I thought was probably right to allow for voltage sag under load.

 

So I connected it up and then measured the 7805 and lo and behold... it's way higher than 5V again... like WTF?

 

This is without the board connected to the computer. I mean surely this board couldn't have the same fault as my other board? Or this is actually correct.

When the board is NOT connected, that pin is delivering more volts. Maybe it shouldn't actually be powered up without having any load?

 

Anyway, I'd really appreciate if anyone can verify what the voltage on pin3 of the 7805 is under these conditions.

[slaanesh]

So what I mean is that when the power board isn’t connected to the rest of the computer, the 7805 is measuring about 10V on pin3 output pin.

The 7805 is on the outside of the heatsink and pin 3 is on the right hand side. Please correct me if I’m wrong.

[TGB1718]

Whats on Pin 1 ?

 

Looking at the circuit, the strange part is the 33 ohm resistor across the input/output

terminals of the 7805 and it may well be that until a certain load is applied to the circuit

then the voltage at pin 3 may be unstable.

 

The only way to test this would be to put a "lowish" (5 to 10 ohms 10 Watt) resistor

across pins 5 and 3. (if you have one, don't touch it when its on, it will be very hot)

 

This will load the circuit depending on the resistor used to 0.5A to 1A which should be more

than enough to test the PSU.

 

 

Edited January 25, 2020 by TGB1718

[slaanesh]

Quote

Whats on Pin 1 ?

 

Original power board:

Pin 1 has 14V input 

Pin 3 has 11.96V output

 

New power board:

Pin 1 has 14.06V input 

Pin 3 has 12.5V output

Edited January 26, 2020 by slaanesh

[MrMartian]

According to the attached schematic, there is a 33ohm resistor as a bypass around the 7805 (R203). This would mean, that without a load I would expect the voltage to be pretty close to the input voltage.

[slaanesh]

14 hours ago, MrMartian said:

According to the attached schematic, there is a 33ohm resistor as a bypass around the 7805 (R203). This would mean, that without a load I would expect the voltage to be pretty close to the input voltage.

 

Okay well that really would exactly explain the readings I am getting.

I'm thinking that both of my power boards are actually fine.

[slaanesh]

So I removed the 48/52K memory board and replaced it with the original Atari 16K board and everything is working again.

The AV mod works fine and the computer boots up every time.

So the power board was fine all along. The new power board is also fine.

 

The issue seems to be with the 48/52K board. I was very careful with all aspects of the mod and I have no reason to believe that I damaged it.

 

Anyway I am getting a replacement board so will see if that fixes the issue and will report back my findings.

[TGB1718]

Nice to see you have it working again, nothing worse than chasing a "red herring", hope the replacement

fixes everything for you too.

 

Please let us know

[BillC]

On 1/26/2020 at 5:59 AM, MrMartian said:

According to the attached schematic, there is a 33ohm resistor as a bypass around the 7805 (R203).

R203 allows the +5V components to draw more current than the 7805 can deliver. It is an inexpensive way to deliver additional current, but the output current must be more than R203 supplies in order to be properly regulated.

(input voltage - output voltage)/33ohm

(10V-5V)/33ohm=151.5mA

[slaanesh]

So I got the replacement power board and it's absolutely no different.

I basically can't get the RAM board to work with my Atari 400.

The guy that does the 48/52K memory boards graciously even sent me a second board which I tested with exactly the same results.

I ended up paying him for the second board anyway as I thought maybe the original was faulty but it wasn't.

 

When I remove the 48/52K memory board everything works 100% all the time.

Putting the 48/52K memory board back in results in a computer that only boots sometimes. Same for both boards.

 

I give up!

 

So it's back to a 16K Atari 400 and I have 2x 48/52K memory boards... and a second power supply... and a second CPU board!

 

The UAV video board and the internal sound amp are the only mods in the computer and it works a treat like this. Great picture!

 

The only thing I haven't tried is removing the UAV mods and see if the 48/52K memory board works reliably without them... but that defeats the purpose for me.

 

If I can get another Atari 400 I will try it again. But I don't really need/want another but if another does come up I'm willing to try again.

Edited March 13, 2020 by slaanesh

[MrMartian]

I have this board in a 400 and it just worked the first time. One thought I had, when you put the board in what are you doing with the ROM on the main board?

[xrbrevin]

sorry to hear that

i saw that tynemouth software in the UK used to sell his own 48k board for the 400 but he had to discontinue it due to compatibility issues. some work fine and others just dont:

http://blog.tynemouthsoftware.co.uk/2018/11/atari-400-48k-internal-ram-upgrade.html

i wonder if it is the same with the type you have?

i'd like to know what causes it because im looking to upgrade my 400 to 48k and it would be useful to know before i commit

[Tea Party]

I had one of the tynemouth boards in my 400 and was getting all sorts of random crashes, graphics corruption, collision detection and other problems including ‘birdies’. 
 

I installed one of tf_hh’s 48/52K boards and all of my problems disappeared straight away. 

[jamm]

On 3/13/2020 at 4:25 AM, slaanesh said:

So I got the replacement power board and it's absolutely no different.

I basically can't get the RAM board to work with my Atari 400.

The guy that does the 48/52K memory boards graciously even sent me a second board which I tested with exactly the same results.

I ended up paying him for the second board anyway as I thought maybe the original was faulty but it wasn't.

 

When I remove the 48/52K memory board everything works 100% all the time.

Putting the 48/52K memory board back in results in a computer that only boots sometimes. Same for both boards.

 

I give up!

 

So it's back to a 16K Atari 400 and I have 2x 48/52K memory boards... and a second power supply... and a second CPU board!

 

The UAV video board and the internal sound amp are the only mods in the computer and it works a treat like this. Great picture!

 

The only thing I haven't tried is removing the UAV mods and see if the 48/52K memory board works reliably without them... but that defeats the purpose for me.

 

If I can get another Atari 400 I will try it again. But I don't really need/want another but if another does come up I'm willing to try again.

If we're talking about tf_hh's upgrade board, then I, unfortunately, had the same issue with mine.  Luckily I tested immediately and reverted the upgrade to identify the problem as the RAM upgrade board.

 

Since it boots correctly sometimes (1/10 of the time in my case), tf_hh thinks it's a timing issue, which makes sense.  We're going to keep trying different things and hopefully get to the bottom of the problem eventually...

 

 

[slaanesh]

 

On 3/13/2020 at 11:13 PM, MrMartian said:

I have this board in a 400 and it just worked the first time. One thought I had, when you put the board in what are you doing with the ROM on the main board?

I've tried a number of things including removing the ROMs from the Atari 400 board and using the 48/52K ROMs and the reverse as well (and setting the jumpers as required).

But to no success. 

 

So at the moment, I am not using the 48/52K board.

 

Is there anyone else (apart from @jamm) who has the tf_hh 48/52K board and it's not working reliably? I mean the actual board is excellent, it's just in combination with my particular Atari 400.

[areeve]

As I'm having issues that this thread deals with I'll throw out that the schematic appears to be labeled incorrectly... the R203 in the schematic that is discussed here is really R202 on the actual power supply board (the real R203 as labeled on the power supply board is close to the bottom middle of the schematic). After doing a 48k memory upgrade and a recap my 400 went to the familiar blue memo pad screen (just blank) and then the system died (power light went out / black screen). In diagnosing I discovered that the +5v from the 7805 was really +13v. My first suspect was the voltage regulator so I pulled that off and it's fine. I then tested the board without it and was still getting +13v on the (empty) output through hole where the 7805 was which is when I noted the resistor could be one culprit. In circuit it was a short so I pulled it to verify and it's a short alright...

 

Thanks for the info on its purpose. I'm still not clear on why I'd expect a +13v reading from a multimeter. We're not saying that if I check the SIO +5v pin that I'd see anything other than +5v are we?

 

My concern is what other damage might have occurred as part of this fault ?

Edited January 21, 2021 by areeve

[areeve]

Actually I am wrong about the resistor... it's just such a low value resistor at 33 ohms this it passes the continuity test on my meter. So I guess we're saying that reading +13v on the +5v line is normal? I'm just plain puzzled by this.

[BillC]

1 hour ago, areeve said:

Actually I am wrong about the resistor... it's just such a low value resistor at 33 ohms this it passes the continuity test on my meter. So I guess we're saying that reading +13v on the +5v line is normal? I'm just plain puzzled by this.

The 33ohm resistor across the 7805 voltage regulator is to supply additional current without using a regulator that can supply more current, if checking the 7805 output without a load there is no current induced voltage drop on the resistor so the voltage reading is high. To check the true 7805 output you need to apply a load, I load tested an XL power supply with a 6V KPR113 flashlight bulb that is rated 0.75A@4.8V.

[+Stephen]

Yeah - but even unloaded, a 7805 should not output 13V.  That's the point of using a regulator (7805) in the first place, right?

[areeve]

1 hour ago, Stephen said:

Yeah - but even unloaded, a 7805 should not output 13V.  That's the point of using a regulator (7805) in the first place, right?

The reading isn't coming from the 7805... as above I removed that and got the same reading. The reading is coming from across the resistor that bypasses the 7805 and is high (on the schematic that is also above), apparently, because of a lack of a load... but we're getting a bit beyond my 'learner' understanding of all this. I guess it sounds like my power supply is fine, though and I need to start digging into the main board and what's going on over there.

[BillC]

1 hour ago, Stephen said:

Yeah - but even unloaded, a 7805 should not output 13V.  That's the point of using a regulator (7805) in the first place, right?

The regulator isn't outputting 13V, the voltage is coming through the 33 ohm resistor, without any current there is no voltage drop across it which is why I suggested a load test.

 

The resistor should deliver about 240mA current over what the regulator can supply,(13V-5V)/33 ohm, as longer as the load requires more current than this the voltage should be properly regulated.

[+Stephen]

2 hours ago, areeve said:

The reading isn't coming from the 7805... as above I removed that and got the same reading. The reading is coming from across the resistor that bypasses the 7805 and is high (on the schematic that is also above), apparently, because of a lack of a load... but we're getting a bit beyond my 'learner' understanding of all this. I guess it sounds like my power supply is fine, though and I need to start digging into the main board and what's going on over there.

 

1 hour ago, BillC said:

The regulator isn't outputting 13V, the voltage is coming through the 33 ohm resistor, without any current there is no voltage drop across it which is why I suggested a load test.

 

The resistor should deliver about 240mA current over what the regulator can supply,(13V-5V)/33 ohm, as longer as the load requires more current than this the voltage should be properly regulated.

This is why I don't do hardware.  Sorry guys.