Hi,
I made some measurements on my KC-7, and saw something strange when I looked at the PS rails.
On the attached picture, you can see how the voltage is at the comparator's supply pins.
The upper curve is the positive supply, the lower curve is the negative supply.
There are 20mV peaks every 100ns, this corresponds to the clock period.
The strange thing is that there's garbage on the positive supply, but the negative one is clean.
When I look how the voltage is at the regulator's output (before the chokes), it's clean on both rails.
Another thing is that the clock draws 14mA on the positive supply, and 6mA on the negative one.
Elso said once that the clock draws 7mA. I don't know if it's 7mA on both, or total. But 14mA is clearly too much.
The multiturn pot's whiper is at 0.64V
Does anyone have an idea why I have this "noise" on the comparator's positive supply?
I made some measurements on my KC-7, and saw something strange when I looked at the PS rails.
On the attached picture, you can see how the voltage is at the comparator's supply pins.
The upper curve is the positive supply, the lower curve is the negative supply.
There are 20mV peaks every 100ns, this corresponds to the clock period.
The strange thing is that there's garbage on the positive supply, but the negative one is clean.
When I look how the voltage is at the regulator's output (before the chokes), it's clean on both rails.
Another thing is that the clock draws 14mA on the positive supply, and 6mA on the negative one.
Elso said once that the clock draws 7mA. I don't know if it's 7mA on both, or total. But 14mA is clearly too much.
The multiturn pot's whiper is at 0.64V
Does anyone have an idea why I have this "noise" on the comparator's positive supply?
Attachments
Hi Bricolo,
Do you have a 0.1 µF decoupling cap directly at pin 1 (+supply) of the comparator?
Do you have a decent clock output?
You did use chokes in the powersupply line? I only advice small ferrite beads.........
Do you have a 0.1 µF decoupling cap directly at pin 1 (+supply) of the comparator?
Do you have a decent clock output?
You did use chokes in the powersupply line? I only advice small ferrite beads.........
Is it the KC7 and if so are you measuring after L1 or L2?
Edit: sorry you do not need to answer. I saw now that you measure at the comparator.
Edit: sorry you do not need to answer. I saw now that you measure at the comparator.
It's important _how_ you measure. Keep the ground cable very close to the comparator. You can also try to twist the ground cable around the probe in order to make the antenna smaller, less loop area.
How about decoupling very close to the comparator? 10 nF ceramic cap? C4, C15, 100 nF (unknown cap) is maybe not sufficient? BTW, is it the model 7?
How about decoupling very close to the comparator? 10 nF ceramic cap? C4, C15, 100 nF (unknown cap) is maybe not sufficient? BTW, is it the model 7?
Elso:
yes, there's a 0.1uF ceramic cap just aside each supply leg of the comparator
the clock output looks OK, you've seen pictures of it on another thread
of course it's ferrite beads, I wrote the wrong word.
Hjelm:
yes, directly on the comparator.
Peranders:
this is how the grounding on my probe is done (like the one on the left)
BTW, since the negative rail is clean, I don't think the problem is in the measurement stuff.
Andypairo:
Thank you.
If you can measure the current drawn from each polarity it would be nice too.
yes, there's a 0.1uF ceramic cap just aside each supply leg of the comparator
the clock output looks OK, you've seen pictures of it on another thread
of course it's ferrite beads, I wrote the wrong word.
Hjelm:
yes, directly on the comparator.
Peranders:
this is how the grounding on my probe is done (like the one on the left)
An externally hosted image should be here but it was not working when we last tested it.
BTW, since the negative rail is clean, I don't think the problem is in the measurement stuff.
Andypairo:
Thank you.
If you can measure the current drawn from each polarity it would be nice too.
Wrong Bead
I am terrible sorry but you got the wrong bead or the right bead at the wrong place. You can use your beads BL02... but before the regulator.😱
You can use Farnell part# 3234861 just before the comparator.
Hi Bricolo,Bricolo said:
I am terrible sorry but you got the wrong bead or the right bead at the wrong place. You can use your beads BL02... but before the regulator.😱
You can use Farnell part# 3234861 just before the comparator.
Beads
Hi,
I am probably using these:
www.elfa.se
Stock number: 58-727-26
Product name: Ferrite bead FB-73-101
Manufacturer: Amidon http://www.amidoncorp.com/aai_ferritesshieldingbeads.htm
Type designation: FB-73-101
Edit:
Actually it is this one:
Ferritebead for HF decoupling. Maker Philips. Outside diameter 3,7mm. Length 3,5mm. Has one hole of 1,2mm diameter.
Sorry no datasheet.
Hi,
I am probably using these:
www.elfa.se
Stock number: 58-727-26
Product name: Ferrite bead FB-73-101
Manufacturer: Amidon http://www.amidoncorp.com/aai_ferritesshieldingbeads.htm
Type designation: FB-73-101
Edit:
Actually it is this one:
Ferritebead for HF decoupling. Maker Philips. Outside diameter 3,7mm. Length 3,5mm. Has one hole of 1,2mm diameter.
Sorry no datasheet.
But what's wrong with mines?
I'd like to know how to select appropriate beads, if I can't find the same as yours
I'd like to know how to select appropriate beads, if I can't find the same as yours
Hello,
I measured the noise at the rails and my oscilloscope gives me a sort of noise floor, quite small and unrelated to the clock frequency for both rails.
Current measurements shows that the v+ rail uses much more current, but looking at the schematic that's perfectly normal.
Cheers
Andrea
I measured the noise at the rails and my oscilloscope gives me a sort of noise floor, quite small and unrelated to the clock frequency for both rails.
Current measurements shows that the v+ rail uses much more current, but looking at the schematic that's perfectly normal.
Cheers
Andrea
Elso Kwak said:Hi Bricolo,
Do you have a 0.1 µF decoupling cap directly at pin 1 (+supply) of the comparator?
Do you have a decent clock output?
You did use chokes in the powersupply line? I only advice small ferrite beads.........
Elso, what will happen if using a choke but not a small bead after the regulator?
Bricolo said:
I took them off a old motherboard, but normally they are made of a piece of wire surrounded with ferrite... nothing really special.
Currents are 16mA and 6mA (from a +/- 12V supply)
The pic shows them.
BTW Does anyone know the brand/type of the grey caps?
Cheers
Andrea
Hint
http://www.diyaudio.com/forums/showthread.php?s=&threadid=33041&perpage=15&highlight=&pagenumber=1
(no more hints)
Professor DR. BUNSEN HONEYDEW has probably an idea.
Or has his assistant Beaker a clue?
http://www.diyaudio.com/forums/showthread.php?s=&threadid=33041&perpage=15&highlight=&pagenumber=1
(no more hints)
Professor DR. BUNSEN HONEYDEW has probably an idea.
Or has his assistant Beaker a clue?
Bricolo said:
Not many knows what this "hook" is good for. Short ground cable is _very_ important sometimes.I look at the figures again. 20 mV ripple isn't bad. The comparator takes 16 mA(?) but at the transitions it can be 100 mA or 1 A even. The question is how much capacitance do you have to have if the capacitor was ideal. How much with a real world cap?
Vripple = (I*t)/C => just put in numbers and see what you have got.
Blicolo,
The negative supply only need to supply the comparator, when the positive one supplies to the comparator + the oscillator, the oscillator draws 8mA seems quite reasonable.
Have you try replace a new 0.1uF and meansure again?
The negative supply only need to supply the comparator, when the positive one supplies to the comparator + the oscillator, the oscillator draws 8mA seems quite reasonable.
Have you try replace a new 0.1uF and meansure again?
- Status
- Not open for further replies.