Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Login


3 Pages123>
Options
View
Go to last post Go to first unread
miero  
#1 Posted : Monday, September 1, 2014 7:20:33 AM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
EDIT: Description of the used workaround: http://twistedpearaudio....----FIXED.aspx#post21552

I'm trying to install Placid HD BP v2.1 and the negative voltage does not work right after power on.

It sometimes gets somehow fixed after 1-10 minutes (while I'm trying to measure voltages), however this is mostly random.

Following voltages are measured without load:
TP_VIN- ... -20.5V
TP_OUT- ... -2.78V
total current ... 183mA
shunted current ... 180mA

TP_VREF- ... N/A ... EDITED: bad initial measurement ... correct value -1.9V

voltage drops on components:
QP1 (CCS-OUT) ... 18V (too high) ---> later 8.8V
QP2 (SHUNT-GND)... 2.58V (too low) ---> later 11.7V
R3B ... 18.6V
R5B ... 1.67V
R6B ... 1.0V ---> later 9.1V
R7B ... 0.03V ---> later 0.3V
R8B ... 0.8V
R10B ... 1.3V
R11B ... 1.21V ---> 1.26V
R12B ... 1.9V ---> later 9.4V
R13B ... 0.22V ---> later 0V
VR1B ... 0.2V
VR2B ... 0.78V ---> later 2.48V

Magic follows: Sometimes it gets fixed by measuring voltage on R11B, but only when Legato is not connected. Then having there -12V and connecting load to it the voltage fall backs to -2.8V; and it does not get recovered back to -12V after disconnecting the load.

What could be wrong? Thanks!

Image: https://docs.google.com/...OZ4isAcAQaEhpZlMzM2NLWEE

Edited by user Saturday, September 20, 2014 1:14:35 PM(UTC)  | Reason: add image; clarifications; deleted bad measurement of TP_REF-

avr300  
#2 Posted : Monday, September 1, 2014 6:27:07 PM(UTC)
avr300

Rank: Member

Groups: Member
Joined: 6/17/2008(UTC)
Posts: 921
Denmark

Thanks: 1 times
Was thanked: 70 time(s) in 69 post(s)
2 things comes to my attention.

1. Bad solder joint although yours look fine. Better do an extra check.
2. Those offboard mounted darlingtons. Are their leg extensions introducing instability?

Do you have a scope?
miero  
#3 Posted : Tuesday, September 2, 2014 5:36:27 PM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
It seems stable and it does not oscillate. However it moves to different stable state than it should.

Here you can see a start on the negative output; TP_SHUNT- is red, TP_OUT- is green:
miero attached the following image(s):
placid-neg-start--shunt-and-out.png (30kb) downloaded 18 time(s).

You cannot view/download attachments. Try to login or register.
Brian Donegan  
#4 Posted : Tuesday, September 2, 2014 6:39:22 PM(UTC)
Brian Donegan

Rank: Administration

Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 2,868
Location: Massachusetts, USA

Thanks: 2 times
Was thanked: 141 time(s) in 134 post(s)
Can you post a picture of the bottom of the board? My first guess is also cold joint.

Another test would be to check the resistance of the trimmers over time (power off) while poking around the board with a finger. See if they fluctuate at all.

miero  
#5 Posted : Tuesday, September 2, 2014 9:53:13 PM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
I've re-soldered everything suspicious and the PCB has already started wearing out... here is bottom of negative part of the board: https://drive.google.com...OZ4isAcAQckxTNmtOMkZaOWc

The problem still persists...

In this "alternative mode" the VR2B changes voltage also on the TP_VREF- and even on the maximum position the TP_VREF- is not what it should be... QP2 shunts much more than is needed.

EDIT: new pic after cleaning R11B lead and re-soldering wires to QP1 and QP2... https://drive.google.com...OZ4isAcAQU2VGNjZaQURlX0k

Edited by user Tuesday, September 2, 2014 10:26:39 PM(UTC)  | Reason: Not specified

miero  
#6 Posted : Friday, September 5, 2014 8:05:03 AM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
EDIT: Initially I had incorrectly rewritten schematics into simulation. So I've deleted my false statements to not confuse readers.

Here are part of simulation circuit I've prepared during debugging of my Placid... it is still not fixed :-(

Edited by user Friday, September 5, 2014 6:31:40 PM(UTC)  | Reason: Not specified

File Attachment(s):
placid21negshunt.fixed3.zip (7kb) downloaded 5 time(s).
miero attached the following image(s):
placid3.png (33kb) downloaded 23 time(s).

You cannot view/download attachments. Try to login or register.
miero  
#7 Posted : Saturday, September 6, 2014 12:46:56 AM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
I've desoldered QP2 and set the low current on the VR1B. After powering on the "reduced Placid" is able to control output voltage just using QN4.

Thus problem might be either damaged QP2 or increased capacitance on wires to QP2, because extra capacitance breaks also simulation with OPA209 if added between base and collector... or something else...

Edited by user Saturday, September 6, 2014 12:48:13 AM(UTC)  | Reason: Not specified

miero  
#8 Posted : Saturday, September 6, 2014 11:16:52 AM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
It will be something else... :-|

I replaced both QP1 and QP2 on wires with brand new Qs TIP127 connected directly to PCB and result is the same ... -2.8V at negative output after powering Placid 2.1 on.

After it is on with the -2.8V at negative output following works to fix output voltage: disconnect QP2 and connect it again. After that it runs well... (tested with limited current only)

Edited by user Saturday, September 6, 2014 11:17:43 AM(UTC)  | Reason: Not specified

miero  
#9 Posted : Saturday, September 6, 2014 1:02:43 PM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
Any ideas? What else could be wrong?
avr300  
#10 Posted : Sunday, September 7, 2014 8:18:50 AM(UTC)
avr300

Rank: Member

Groups: Member
Joined: 6/17/2008(UTC)
Posts: 921
Denmark

Thanks: 1 times
Was thanked: 70 time(s) in 69 post(s)
"TP_VREF- ... -2,78V"

hmm, it should be -2.50v
miero  
#11 Posted : Sunday, September 7, 2014 10:36:00 PM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
-

Edited by user Monday, September 8, 2014 6:56:02 AM(UTC)  | Reason: deleted false claims based on incorrect measurement of TP_REF-

miero  
#12 Posted : Monday, September 8, 2014 7:05:20 AM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
avr300: Thanks for noticing, I've checked TP_REF- again and it was different that wrote previously!

Measurements with TIP127 darlingtons:

TP_REF- ... -1.89V
R6B ... other leg: -3.15V ... v.drop.: 1.26V ... current: 0.63mA
R13B ... other leg: -1.6V ... v.drop.: 0.29V ... current: 0.62mA
IC2 in+ ... -1.6V
IC2 in- ... -0.73V
IC2 out ... -2.39V
IC2 Vs+ ... 0V
IC2 Vs- ... -3.15V
TP_OUT- ... -3.18V
VR2B ... 392 Ohm

Also I've checked it on 100MHz scope today and voltage on IC2 inputs/outputs is stable.

Edited by user Monday, September 8, 2014 9:53:13 AM(UTC)  | Reason: Not specified

miero  
#13 Posted : Monday, September 8, 2014 11:08:54 AM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
Isn't 0.62mA current on R13B (IC2 in+) strangely high? The input impedance of OPA209 is 200kOhm.

Could a C8B be "leaky"?

Datasheet for that capacitor EEU-FC1H470H specifies leakage current(max.) is 23.5uA.

So I'm going to try replace C8B...

Edited by user Monday, September 8, 2014 11:17:40 AM(UTC)  | Reason: Not specified

miero  
#14 Posted : Monday, September 8, 2014 6:11:13 PM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
Similar situation with removed C8B... it must be something else...
avr300  
#15 Posted : Monday, September 8, 2014 6:28:22 PM(UTC)
avr300

Rank: Member

Groups: Member
Joined: 6/17/2008(UTC)
Posts: 921
Denmark

Thanks: 1 times
Was thanked: 70 time(s) in 69 post(s)
Please measure R6B and R7B.

Are they both within the expected values?
miero  
#16 Posted : Monday, September 8, 2014 8:57:14 PM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
Matches as the values from the first post:

Voltage drop on resistors:
R6B ... 1.0V ---> later 9.28V
R7B ... 0.03V ---> later 0.3V

"Later" is after disconnecting and connecting QP2.
miero  
#17 Posted : Monday, September 8, 2014 9:29:38 PM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
Oh... some progress... I've attached QP2 to longer wires, connected wires to leads of R10 and R2 and also added a switch to disconnect its pin 1 ... and now it works... eh. (EDIT: because used switch has 8.4kOhm resistance even when it is switched off).

Is that due to added capacitance? (EDIT: partially, because capacitors might not have been always fully discharged and voltage reference was still at 2.5V -- turning on in that case resulted in the good behavior)

(EDIT: deleted text about switch at pin 2)

More tomorrow...

Edited by user Tuesday, September 9, 2014 8:51:20 AM(UTC)  | Reason: Not specified

Russ White  
#18 Posted : Tuesday, September 9, 2014 2:45:19 AM(UTC)
Russ White

Rank: Administration

Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN

Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
Switch? What?

It likely not capacitance at all....

Why the wires at all? if you use shorter stand offs you could mount directly to the pads.
miero  
#19 Posted : Tuesday, September 9, 2014 7:13:48 AM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
When QP2 is soldered directly it gets opened too soon and IC2 finds alternative stable state at -2.8V.

Here is what I've meant with "the switch".

Delaying open of QP2 "fixes" the problem.

It's no solution, just my silly hack to find out what is wrong with this twisted power supply.
miero attached the following image(s):
placid-cudofix.png (39kb) downloaded 24 time(s).

You cannot view/download attachments. Try to login or register.
miero  
#20 Posted : Tuesday, September 9, 2014 7:19:38 AM(UTC)
miero

Rank: Member

Groups: Member
Joined: 10/22/2011(UTC)
Posts: 135
Location: Prague

Thanks: 1 times
Was thanked: 3 time(s) in 3 post(s)
With the real switch it looks even more scary... :D

https://drive.google.com...ckVnZm8/edit?usp=sharing

Edited by user Tuesday, September 9, 2014 8:44:19 AM(UTC)  | Reason: Not specified

Rss Feed  Atom Feed
Users browsing this topic
GuestUser (2)
3 Pages123>
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.