Hey Guys...
I have yet another quick question.
Lets say for example in the bias supply, I have a LCLC supply as indicated a few posts back, and at the end is a potential divider (10K resistor-10Kpot -10K resistor) which is used to fine-tune the output voltage. When the grid draws current, is the voltage going to change on the grid as a function of the resistances in the potential dividers? That is, does the resistance in the potential divider upstream of the pot act similarly as the resistance in a CRC filter? I assume it would as I am still drawing current through the resistor thus causing voltage drop, right?
Thanks,
BK
I have yet another quick question.
Lets say for example in the bias supply, I have a LCLC supply as indicated a few posts back, and at the end is a potential divider (10K resistor-10Kpot -10K resistor) which is used to fine-tune the output voltage. When the grid draws current, is the voltage going to change on the grid as a function of the resistances in the potential dividers? That is, does the resistance in the potential divider upstream of the pot act similarly as the resistance in a CRC filter? I assume it would as I am still drawing current through the resistor thus causing voltage drop, right?
Thanks,
BK
Hi, sorry for the delayed reply, I have been busy. The current drawn through the grid circuit, and thus from the bias supply is all provided by the driver tube IF (a big if) the bias supply has a low impedance for all audio frequencies. Since music is not a sine wave and you will likely not be operating the amp in continuous A2 (only on musical peaks) the DC resistance of the power supply does not need to be extremely low. If you use a simple voltage divider, the capacitor must provide the low impedance to ground for all audio frequencies.
When I was experimenting with an IT I used a pot feeding a source follower made with a P channel mosfet to provide a low impedance buffer for the bias supply. I made the circuit up as I went along. I know that I scratched the schematic down somewhere, but I can't find it. You would need 2 of these, one for each channel.
It was something rediculously simple, drain tied to the B- supply, bias pot connected to the gate, resistor and cap connected from source to ground. The source is the feed point to the transformer. The resistor was big, like 10K 10 watt, but it diidn't get hot, it was what I could find at the time.
When I was experimenting with an IT I used a pot feeding a source follower made with a P channel mosfet to provide a low impedance buffer for the bias supply. I made the circuit up as I went along. I know that I scratched the schematic down somewhere, but I can't find it. You would need 2 of these, one for each channel.
It was something rediculously simple, drain tied to the B- supply, bias pot connected to the gate, resistor and cap connected from source to ground. The source is the feed point to the transformer. The resistor was big, like 10K 10 watt, but it diidn't get hot, it was what I could find at the time.
I found the schematic!
I used this with a push pull transformer driven pentode amp experiment. Some values may need to be changed. The schematic shows a zener diode. You could use a string of zeners or VR tubes to make this into a bias filter and regulator. Leave these parts out for a filter - buffer only circuit.
There is good reason NOT to use a bias regulator if you are not regulating the plate voltage in a triode amp. This will make the plate current very dependent on the power line voltage. In a pentode amp with a screen regulator, it is good to regulate the bias voltage.
I used this with a push pull transformer driven pentode amp experiment. Some values may need to be changed. The schematic shows a zener diode. You could use a string of zeners or VR tubes to make this into a bias filter and regulator. Leave these parts out for a filter - buffer only circuit.
There is good reason NOT to use a bias regulator if you are not regulating the plate voltage in a triode amp. This will make the plate current very dependent on the power line voltage. In a pentode amp with a screen regulator, it is good to regulate the bias voltage.
Attachments
Hey Hey Guys,
So I need some input.
As mentioned in a few posts back, I am having problems powering up the amp when connected to the mains w/o the variac.
As mentioned in the previous posts, the 2A fuse on the mains was not up to the task. So, I decided to repeat this with a 2.5A fast blow. When the delay relay triggered, the fuse blew as observed before (expected result). That said, I then switched to the 4A slow-blow.
When the mains was switched, the fils, regulator cooling fan, and input/driver rectifier heater all powered up w/o a problem. After about a 90 second delay, the relay triggered and applied power the the input/driver B+ and the 845 B+ circuit (6AU4 heaters and HV txfmr). The B+ to the driver came up immediately as expected and the 6AU4s began to warm up. I had the DVM on the stack of bleeder resistors for the 845 B+ filter caps as a means to measure B+. After about 10 seconds the voltage starts to climb, and after another 10 seconds starts to raise at a quick rate. When the DVM measured about 400V (this is at the juntion between the second and 3rd resistor = ~2/3 B+) the fuse blew as before...
So, what I noticed as the fuse blew was an arc in the 6AU4 and a rumble (mechanical noise) which came from the 6AU4s. Also, the B+ on the filter caps was very quickly discharged, where as it usually takes a few minutes to bleed down, in this case it was down in the teens between the time I first say the arc in the 6AU4 and the time it took me to look over at the DVM (a couple seconds). So, what is going on here. Could I have a bad 6AU4, or is the 33uF-10H-100uF-10H-100uF filter causing too much current to be drawn through the 6AU4s thus resulting in the arc/rumble?
I may try to bring up the 845B+ with the variac again to ensure the circuit is intact. Everything checks fine, but the problem persists. I'll await advice before I continue.
Thoughts?
So I need some input.
As mentioned in a few posts back, I am having problems powering up the amp when connected to the mains w/o the variac.
As mentioned in the previous posts, the 2A fuse on the mains was not up to the task. So, I decided to repeat this with a 2.5A fast blow. When the delay relay triggered, the fuse blew as observed before (expected result). That said, I then switched to the 4A slow-blow.
When the mains was switched, the fils, regulator cooling fan, and input/driver rectifier heater all powered up w/o a problem. After about a 90 second delay, the relay triggered and applied power the the input/driver B+ and the 845 B+ circuit (6AU4 heaters and HV txfmr). The B+ to the driver came up immediately as expected and the 6AU4s began to warm up. I had the DVM on the stack of bleeder resistors for the 845 B+ filter caps as a means to measure B+. After about 10 seconds the voltage starts to climb, and after another 10 seconds starts to raise at a quick rate. When the DVM measured about 400V (this is at the juntion between the second and 3rd resistor = ~2/3 B+) the fuse blew as before...
So, what I noticed as the fuse blew was an arc in the 6AU4 and a rumble (mechanical noise) which came from the 6AU4s. Also, the B+ on the filter caps was very quickly discharged, where as it usually takes a few minutes to bleed down, in this case it was down in the teens between the time I first say the arc in the 6AU4 and the time it took me to look over at the DVM (a couple seconds). So, what is going on here. Could I have a bad 6AU4, or is the 33uF-10H-100uF-10H-100uF filter causing too much current to be drawn through the 6AU4s thus resulting in the arc/rumble?
I may try to bring up the 845B+ with the variac again to ensure the circuit is intact. Everything checks fine, but the problem persists. I'll await advice before I continue.
Thoughts?
If you could post schematic...or I will go back and look in older posts....
Start-up of a power supply and circuit need to be sequenced properly... SOunds like the 6AU4 might have shorted....
You may need to put in small zeners to hold the voltages in place before the amp is up and properly pulling currents... I won't know till I see the circuit...
Chris
Start-up of a power supply and circuit need to be sequenced properly... SOunds like the 6AU4 might have shorted....
You may need to put in small zeners to hold the voltages in place before the amp is up and properly pulling currents... I won't know till I see the circuit...
Chris
As mentioned, everything is powered up with the mains switch with the exception of the 845 B+ supply (HV txfmr and 6AU4's) and the input/driver HV txfmr.
When I run the sim in PSUD I do get a warning about exceednig the peak current of the diode... Maybe this is an indication of the problem.
Thoughts?
When I run the sim in PSUD I do get a warning about exceednig the peak current of the diode... Maybe this is an indication of the problem.
Thoughts?
Tubelab,
I thought about doing that as a test to determine excessive current draw. The only problem with that is should it work, and I need to drasitcally downsize the caps, my target 1000VDC will probably be out of range with the current transformer.
Either way, I will give it a go with the choke input AND variac. If that works, I will bring it up without the variac. If all is good, I guess I can assume the input cap is just too big!
I will also run a few sims in PSUD to see at what point I need to go bigger with the transformer. If all else fails, and I need to have Jack wind another txfmr, I guess I could have him wind me a 2000VCT and go choke input... I'll report back what I find later tonight.
Does arcing over in the diode cause any permenant damage that I should be concerned about???
If anything else pops into your mind(s) feel free to chime in.
Thanks,
BK
I thought about doing that as a test to determine excessive current draw. The only problem with that is should it work, and I need to drasitcally downsize the caps, my target 1000VDC will probably be out of range with the current transformer.
Either way, I will give it a go with the choke input AND variac. If that works, I will bring it up without the variac. If all is good, I guess I can assume the input cap is just too big!
I will also run a few sims in PSUD to see at what point I need to go bigger with the transformer. If all else fails, and I need to have Jack wind another txfmr, I guess I could have him wind me a 2000VCT and go choke input... I'll report back what I find later tonight.
Does arcing over in the diode cause any permenant damage that I should be concerned about???
If anything else pops into your mind(s) feel free to chime in.
Thanks,
BK
Momentary arcing usually does not cause permanent damage unless there was a continuous arc. That could cause enough localized heating to break down the heater to cathode insulation.
I get sparks in the 5AR4 in one of my amplifiers every so often. It uses a 100uf input cap which is way to large. It was the only cap that I had at the time. It still has the same tube after 2 1/2 years.
If the removal of the cap stops the fireworks, you can try a smaller value cap there. It may be possible to find a cap that gives you enough voltage without fireworks.
You don't show the heater winding for the 6AU4's. Is it grounded?
This could cause a heater to cathode breakdown, resulting in fireworks.
Another possibility is to use SS rectification with HV FRED diodes. These have a much higher peak current rating.
I get sparks in the 5AR4 in one of my amplifiers every so often. It uses a 100uf input cap which is way to large. It was the only cap that I had at the time. It still has the same tube after 2 1/2 years.
If the removal of the cap stops the fireworks, you can try a smaller value cap there. It may be possible to find a cap that gives you enough voltage without fireworks.
You don't show the heater winding for the 6AU4's. Is it grounded?
This could cause a heater to cathode breakdown, resulting in fireworks.
Another possibility is to use SS rectification with HV FRED diodes. These have a much higher peak current rating.
Tubelab,
WRT heater winding... Off the top of my head, I cant remember. I use seperate txfmrs for the heater and the HV. The HV txfmr is a 1500VCT and the CT is grounded. The Supply filter network is also grounded through an an I additional wire coming off the negative of the first cap.
The heater txfmr is a 6.3 VCT and I believe as it is currently set up the heater is floating, but I will check on this in the evening when I am in front of the beast. Would this be the correct arrangement, floating the heater or should the CT be referrenced to ground?
WRT heater winding... Off the top of my head, I cant remember. I use seperate txfmrs for the heater and the HV. The HV txfmr is a 1500VCT and the CT is grounded. The Supply filter network is also grounded through an an I additional wire coming off the negative of the first cap.
The heater txfmr is a 6.3 VCT and I believe as it is currently set up the heater is floating, but I will check on this in the evening when I am in front of the beast. Would this be the correct arrangement, floating the heater or should the CT be referrenced to ground?
Your filament transformer winding should be floating, and the transformer should be rated for at least 1500 volts insulation. These transformers are more expensive than a common filament transformer, so many builders use a conventional filament transformer and ground one side (or the CT) of the secondary. This is OK in most amplifiers, but the 6AU4 is good for 900 volts H-K and you have 1000.
Tubelab,
So let me get this straight....
1. The filament should be floating. If it is floating, do we still excede the 900 H-K rating of the 6AU4? With it referenced to earth, it is clear that the H-K would be 1000V as ytou mention.
Now, if I am exceding this rating, what tis the outcome. Does the potential on the K (C?) jump (arc) ;to the heater winding and thus short?
If the heater is floating, what would the approximate H-K be, or is this something to be measured?
Lastly, if I am exceding the H-K limit of the tube, would/should I reference the heater to a higher potential derived from the driver supply or contemplate different rectification options.....
BK
So let me get this straight....
1. The filament should be floating. If it is floating, do we still excede the 900 H-K rating of the 6AU4? With it referenced to earth, it is clear that the H-K would be 1000V as ytou mention.
Now, if I am exceding this rating, what tis the outcome. Does the potential on the K (C?) jump (arc) ;to the heater winding and thus short?
If the heater is floating, what would the approximate H-K be, or is this something to be measured?
Lastly, if I am exceding the H-K limit of the tube, would/should I reference the heater to a higher potential derived from the driver supply or contemplate different rectification options.....
BK
In the ham radio transmitter days of the past we tied the heater winding directly to the cathode of the tube. Often the rectifiers were DHT's (866's) so this was always the case. This puts the full B+ on the transformer secondary. The transformers had to be rated to operate at this voltage. Since no one builds kilowatt ham transmitters anymore, and few people build high power tube audio, these transformers have become scarce and expensive.
Using damper diodes gives you some flexibility. the H-K breakdown voltage is 900 volts if the heater is NEGATIVE with respect to the cathode. Clearly a grounded heater would violate this rating. Many people simply leave the heater winding floating, however this is not always a good idea, since leakage currents can cause this point to wander around. I would tie the secondary CT to a voltage divider so that about half of the B+ is on the secondary. This would put 500 volts across the tube, and 500 volts across the transformer. Both should be happy at these voltages.
I doubt that this has anything to do with your problem since you would need a shorted tube AND a shorted transformer to blow the fuse. This is an unlikely possibility.
Exceeding the H-K breakdown voltage can cause a short or leakage from the heater to the cathode. This used to happen fairly often in the early TV sets. As the tube manufacturers figured out how to make better tubes, this problem went away. A damper diode (6AU4 and others) was designed to operate with a high voltage on the cathode (boosted B+) and the filament grounded or connected DIRECTLY to the power line! They evolved into some pretty tough tubes that are hard to kill.
Using damper diodes gives you some flexibility. the H-K breakdown voltage is 900 volts if the heater is NEGATIVE with respect to the cathode. Clearly a grounded heater would violate this rating. Many people simply leave the heater winding floating, however this is not always a good idea, since leakage currents can cause this point to wander around. I would tie the secondary CT to a voltage divider so that about half of the B+ is on the secondary. This would put 500 volts across the tube, and 500 volts across the transformer. Both should be happy at these voltages.
I doubt that this has anything to do with your problem since you would need a shorted tube AND a shorted transformer to blow the fuse. This is an unlikely possibility.
Exceeding the H-K breakdown voltage can cause a short or leakage from the heater to the cathode. This used to happen fairly often in the early TV sets. As the tube manufacturers figured out how to make better tubes, this problem went away. A damper diode (6AU4 and others) was designed to operate with a high voltage on the cathode (boosted B+) and the filament grounded or connected DIRECTLY to the power line! They evolved into some pretty tough tubes that are hard to kill.
Update...
First, the 6AU4 heaters txfmr is floating. They are wired in series, with both legs of the secondary feeding one tubes heater (pin 7 and 8) and then again to the next tube (pin 7 and 8). The secondaries from the HV txfmr then feed pins 5. This looks fine to me.
So, as discussed earlier today, I wanted to try bringing up the 845 B+ with the variac nice and slow...., just to see that the curcuit is intact. Interestingly, when the B+ was at about 300 volts there was an arc on one (I think only one) of the 6AU4s. The arc happened in the middle of the tube and seemed to be directed from the inside of the tube towards the outside of the tube... I backed off, and then brought up the voltage again. This time the arc happened at about 500V. Backing off, and bringing up the voltage again It happened at 650V. I stopped here.
So, would this lead you to believe that the tube is bad, or could this mean something more serious? I would like some expertice before I continue...
Thoughts?
First, the 6AU4 heaters txfmr is floating. They are wired in series, with both legs of the secondary feeding one tubes heater (pin 7 and 8) and then again to the next tube (pin 7 and 8). The secondaries from the HV txfmr then feed pins 5. This looks fine to me.
So, as discussed earlier today, I wanted to try bringing up the 845 B+ with the variac nice and slow...., just to see that the curcuit is intact. Interestingly, when the B+ was at about 300 volts there was an arc on one (I think only one) of the 6AU4s. The arc happened in the middle of the tube and seemed to be directed from the inside of the tube towards the outside of the tube... I backed off, and then brought up the voltage again. This time the arc happened at about 500V. Backing off, and bringing up the voltage again It happened at 650V. I stopped here.
So, would this lead you to believe that the tube is bad, or could this mean something more serious? I would like some expertice before I continue...
Thoughts?
OK, problem solved!
It seems that while testing, I have been attaching my DVM probe to the string of three 390K 2 watt resistors which make up the bleeders for each of the 845 B+ filter caps. After doing this several times, the resistors were pulled down towards the body of the cap. Well, it seems that the junction of the resistor was close enough to the body of the cap to arc over. I guess as the short happened, it instantaneously drew excesive current in the rectifier causing the observed arc in the diode.
I bent up the resistors, removing their junction from the cap body, and eveything is powering up just as expected. I'm getting 1000VDC with 70mA through the 845.
One thing I dont like is that I measure the B+ between the 2nd and 3rd resistor in the bleeders (1000VDC is the limit of my meter). I think just to avoid the spook factor, I will measure between the first and second from now on... I don't like the reading otherwise....
It seems that while testing, I have been attaching my DVM probe to the string of three 390K 2 watt resistors which make up the bleeders for each of the 845 B+ filter caps. After doing this several times, the resistors were pulled down towards the body of the cap. Well, it seems that the junction of the resistor was close enough to the body of the cap to arc over. I guess as the short happened, it instantaneously drew excesive current in the rectifier causing the observed arc in the diode.
I bent up the resistors, removing their junction from the cap body, and eveything is powering up just as expected. I'm getting 1000VDC with 70mA through the 845.
One thing I dont like is that I measure the B+ between the 2nd and 3rd resistor in the bleeders (1000VDC is the limit of my meter). I think just to avoid the spook factor, I will measure between the first and second from now on... I don't like the reading otherwise....
In my 845, I measure the B+ from GROUND to the junction of the first and second bleeder resistor. There are 3. If you want to be exact you can measure the value of each resistor (disconnected from the circuit) and compute the exact voltage. I being lazy, just multiply the reading by 3. My meter claims to handle 1000 volts, but the leads look like they might take 400 or so volts. This keeps the whole meter circuit below 400 volts. It is a $5 meter, I don't want to trust it with my life!
Glad to see that the problem turned out to be easy. Sometimes problems with HV circuits are hard to find, other times the trail of smoke and burnt parts makes it obvious.
Glad to see that the problem turned out to be easy. Sometimes problems with HV circuits are hard to find, other times the trail of smoke and burnt parts makes it obvious.
Tubelab,
Thanks for the words of wisdom. I too now read from ground to the first resistor. When I read to the second, the reading was 666, and I didn't like that too much.
Fortunately I have no burnt parts, and the problem was fairly obvious once it became obvious!
The lesson learned here is to be EXTRA careful, and to double check EVERYTHING after fiddling with the circuit.
Anyway, she powers up just fine. I still don't have a proper dummy load, so I am waiting for it's arival before I see if she makes any noise. I'm quite excited to run some wave-forms through her ala pink floyd!
Quick question. What happens when we run the amp without a load on the OPT? How long does it take to damage the opt when run like this?
Thanks,
BK
Thanks for the words of wisdom. I too now read from ground to the first resistor. When I read to the second, the reading was 666, and I didn't like that too much.
Fortunately I have no burnt parts, and the problem was fairly obvious once it became obvious!
The lesson learned here is to be EXTRA careful, and to double check EVERYTHING after fiddling with the circuit.
Anyway, she powers up just fine. I still don't have a proper dummy load, so I am waiting for it's arival before I see if she makes any noise. I'm quite excited to run some wave-forms through her ala pink floyd!
Quick question. What happens when we run the amp without a load on the OPT? How long does it take to damage the opt when run like this?
Thanks,
BK
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.