PH104 said:
Y Dr. Phil – I feel both honored by your invitation and a little humbled to learn that you are under the impression that I would take a cheap shot at you screwing around while holding your nuts with a pair of hemostats. It seems to me that Jen would have been glad to give you a hand with that job.
BTW – a little trick the NASCAR kids use when they are changing tires during pit stops – you can temporarily glue the your nuts in place using an adhesive that will “break away” when the nut is grabbed by the air gun. I’ve never used an air wrench to get my nuts off so I’m unsure as to the proper glue to use - I’ll see if I can find out what the NASCAR kids use for that little trick.
Sorry that I haven’t been around much lately but Susan and I are still trying to get her medical situations under control. I have a new email address that I’ll send around – but you can contact me via the diyAudio email if ya need to.
Here is a link on how to make your own thermal epoxy - http://www.overclock.net/faqs/79115-how-make-your-own-thermal-adhesive.html
lgreen said:
The difference in voltage here is determined by the tolerance spread of the zeners. It will rarely be exactly the same, but always close enough, unless you have two different types of zener.
Phil is correct, 5% will not cause problems.
more tests...
OK, I've got some small heatsinks kludged onto the TO-220 transistors (except Q8 and Q11).
They really don't appear to be getting all that hot so these tiny clip ons appear to be doing the job. Thanks for the thermal epoxy recepie, I used it to mix some AS5 with epoxy but only had to use it on 2 sinks. The rest are clip-ons that I had to hack up and they are not pretty but appear to be working. Not sure if the expoxy I used was "high temp" or if you even need a special type of epoxy. Might come off over time?
The bias pot is turned all the way up (clockwise) and bias starts at 14.6 mV cold/ start on but over time (about 10 min) drifts down to 7.8 mV.
The DC offset pot is turned all the way ccw, offset is about 240 mV and turning it cw increases the offset till the very end (>400 mV) where it starts to decline a little bit but you run out of room to adjust.
Neither pot is a multiturn, I used the part numbers specified by Steve, but not having a multiturn is not the problem since you have good resolution when you turn.
I have turned the variac until I get exactly 50 vdc on the input/ the regulated voltages are +53.8 and -55.5.
R16 has 53.8 on one side and 53.3 on the other.
With power off and measuring the resistances in-circuit, R5 is 53.5 ohms and R14 is 56.7 ohms.
If there is something I should measure, let me know and I'll post.
There is no way to get any measurements on anything near a TO-220 (cannot probe R15, R16, R17, R5, R11) due to the heatsinks covering up the resistors.
Image of Krill with heat sinks (large)
OK, I've got some small heatsinks kludged onto the TO-220 transistors (except Q8 and Q11).
They really don't appear to be getting all that hot so these tiny clip ons appear to be doing the job. Thanks for the thermal epoxy recepie, I used it to mix some AS5 with epoxy but only had to use it on 2 sinks. The rest are clip-ons that I had to hack up and they are not pretty but appear to be working. Not sure if the expoxy I used was "high temp" or if you even need a special type of epoxy. Might come off over time?
The bias pot is turned all the way up (clockwise) and bias starts at 14.6 mV cold/ start on but over time (about 10 min) drifts down to 7.8 mV.
The DC offset pot is turned all the way ccw, offset is about 240 mV and turning it cw increases the offset till the very end (>400 mV) where it starts to decline a little bit but you run out of room to adjust.
Neither pot is a multiturn, I used the part numbers specified by Steve, but not having a multiturn is not the problem since you have good resolution when you turn.
I have turned the variac until I get exactly 50 vdc on the input/ the regulated voltages are +53.8 and -55.5.
R16 has 53.8 on one side and 53.3 on the other.
With power off and measuring the resistances in-circuit, R5 is 53.5 ohms and R14 is 56.7 ohms.
If there is something I should measure, let me know and I'll post.
There is no way to get any measurements on anything near a TO-220 (cannot probe R15, R16, R17, R5, R11) due to the heatsinks covering up the resistors.
Image of Krill with heat sinks (large)
lgreen,
With the descriptions you have given of the problems, it sounds like the trim pots are swapped. R26 should be 1K and R27 10K. What is the bias at the other extreme of pot travel?
With the descriptions you have given of the problems, it sounds like the trim pots are swapped. R26 should be 1K and R27 10K. What is the bias at the other extreme of pot travel?
Interesting. I received an E-mail notice that Marinos had had replied to this thread. After 1 3/4 Hr. I still can not see his post. Can others see it?
not me
Steve, I cannot see his post either.
He was probably making fun of me cause he also thinks that I switched the pots?
Actually I triple checked these because the pots look identical and even measured the resistances. I will look at it again.
Steve, I cannot see his post either.
He was probably making fun of me cause he also thinks that I switched the pots?
Actually I triple checked these because the pots look identical and even measured the resistances. I will look at it again.
I posted asking lgreen for photos but then I saw the link ....
and deleted my post.
Lgreen on the contrary .. I wanted to congratulate you for moving forward..😉
and deleted my post.
Lgreen on the contrary .. I wanted to congratulate you for moving forward..😉
thanks
well thanks for keeping watch. As soon as this problem is solved I can finish up in a day or so; the rest is all done.
ok, I measure 9.5K between #1 and #2 going to the diodes with the diodes connected. I do believe that this is where the resistor is exposed and supposed to be 10K. This drops down to zero when turning the pot. And 9.5K is the value where bias is maximum.
So they do not appear to be swapped.
EDIT- Steve, As I recall, at the other end of travel there is no bias that I can meaure. I will confirm...
well thanks for keeping watch. As soon as this problem is solved I can finish up in a day or so; the rest is all done.
ok, I measure 9.5K between #1 and #2 going to the diodes with the diodes connected. I do believe that this is where the resistor is exposed and supposed to be 10K. This drops down to zero when turning the pot. And 9.5K is the value where bias is maximum.
So they do not appear to be swapped.
EDIT- Steve, As I recall, at the other end of travel there is no bias that I can meaure. I will confirm...
It sounds like they're not swapped. Check the value of R6. Also check the voltage across R12, R14, R16, R17. The value you measure across R5 should produce a negative offset voltage. Check the offset at the end of R11 nearest Q5. How does it change when you adjust R26? Check to confirm the feed-through near C4 is good.
There is clearly a problem here. We will find it.
There is clearly a problem here. We will find it.
Looking at the pcb with the outputs on the left.
R12 is -55.0 on top and -55.6 on the bottom
R14 is -54.7 on top and -55.3 on the bottom
R16 is 53.3 onthe left and 53.0 on the right
R17 is -55.7 on the right and I cannot get to the left.
The PS may have been changing a little during these measurements.
Measuring R11 at the top as suggested. The DC value ranges from 7.4 mV to 90 mV as I turned the offset pot clockwise. It did not go through zero or get negative, just goes up/down.
Also, when cold this could get about 40mV of bias across the emitter resistors, which dropped to 7.6 after about 10 min. I can confirm that when the bias pot is all the way ccw there is no voltage across the emitter resistors.
R12 is -55.0 on top and -55.6 on the bottom
R14 is -54.7 on top and -55.3 on the bottom
R16 is 53.3 onthe left and 53.0 on the right
R17 is -55.7 on the right and I cannot get to the left.
The PS may have been changing a little during these measurements.
Measuring R11 at the top as suggested. The DC value ranges from 7.4 mV to 90 mV as I turned the offset pot clockwise. It did not go through zero or get negative, just goes up/down.
Also, when cold this could get about 40mV of bias across the emitter resistors, which dropped to 7.6 after about 10 min. I can confirm that when the bias pot is all the way ccw there is no voltage across the emitter resistors.
Image
If this will help, here is a very high resolution image of the board on which the problems arise, prior to adding the heat sinks. I take these prior to assembly to review in case things don't work out. It will allow everyone to see what is located where. If you can read color codes you can see if any resistors are messed up. At least you can see if the diodes are pointing in the right direction...
Yes, I am reviewing myself to see if there are issues, and actually measuring the real deal.
- thanks
If this will help, here is a very high resolution image of the board on which the problems arise, prior to adding the heat sinks. I take these prior to assembly to review in case things don't work out. It will allow everyone to see what is located where. If you can read color codes you can see if any resistors are messed up. At least you can see if the diodes are pointing in the right direction...
Yes, I am reviewing myself to see if there are issues, and actually measuring the real deal.
- thanks
All of those resistors should have about 0.6V across them. We don't know about R17, but R16 only has 0.3V across it. Check the voltage across diodes D7,8,9,10. These should all have about 0.6V across them. The voltage across R16 indicates one of the diodes could be shorted.
I've remeasured R16 and it has 0.6 V across it.
the tip of D7 is 52.6/ back is 53.2
the tip of D8 is 53.2/ back is .6 higher/ 53.8
the tip of D9 is -55.6
the tip of D10 is -55.0/ back is .6 higher/ -54.4
Perhaps my regulated supplies are not quite right, how much should they be changing? It seem like every time I do a measurement the supplies are changd a small amount.
the tip of D7 is 52.6/ back is 53.2
the tip of D8 is 53.2/ back is .6 higher/ 53.8
the tip of D9 is -55.6
the tip of D10 is -55.0/ back is .6 higher/ -54.4
Perhaps my regulated supplies are not quite right, how much should they be changing? It seem like every time I do a measurement the supplies are changd a small amount.
lgreen -- do you have a 'scope? It would be useful to look at the output and all of voltage rails. Also, how are you terminating the input when you make the offset measurements?
yes
yes I can hook up the scope. I've got 3 dmms plugged into it now, input is open circ and output is open circ.
yes I can hook up the scope. I've got 3 dmms plugged into it now, input is open circ and output is open circ.
great -- check for oscillation at the output. Also try offset adjustment with the input shorted to signal ground.
Once you nail these problems, I think you'll find the circuit to be very stable and predictable.
Once you nail these problems, I think you'll find the circuit to be very stable and predictable.
Hi,
is it a good general rule to always test with the input loaded?
Either by the active source impedance or a dummy source impedance?
I always do, but that doesn't mean I'm always right.
is it a good general rule to always test with the input loaded?
Either by the active source impedance or a dummy source impedance?
I always do, but that doesn't mean I'm always right.