Hmm, well, it seems the main transformer sags to about 317v. Which is what i would expect based on a lossy tx rated at 250v x ~1.3 (less than root 2) which would be 325v.
I stuck a couple of 5k6 in parallel on the front of my 160v regs. So
B+ supply (317v) ==> parallelled 5k6 5w ==> where it says B+ input on the schematic.
Thats working and showing 317v in and 158v out with about 115v drop accross the paralleled resistors (both 5w so essentially a 10w 2k8) which tallies with my calculations.
However My 260v reg is showing only 220v at its output. Pretty sure this was both sides so fairly sure its not the heater bias dragging it down.
Somethings getting hot, not sure yet what, small amount of smoke but cant see where its coming from and nothing appears to be burnt. It does seem to be coming from the 160v reg somewhere which is a bit odd as I would have thought that having put the extra resistors to drop the input voltage this would have helped.
Could it be that theres only about 40v buffer on that now? I think im getting 198v on the input proper of the 160v reg.
When i tested the psu on its own it didnt hum, but under load it is. Again not sure which tx this yet. Am a bit nervous of poking round too much. Also cant leave it on for very long to track down fault.
Would it be useful to remove the tubes for tracking this down?
The neons seem to be behaving and extinguishing after a few seconds.
I am wondering if my heater tx doesnt like having ~65 v on it but i think i have to have this to avoid damaging the tubes? Not only about reducing hum.
I think i need a clear strategy here to avoid breaking the whole thing or myself.
Would it help if i drew a diagram how ive wired it all up to make sure ive not done anything daft?
Sorry to be a bit vague, switch on time is somewhat limited until I decide on a plan.
On a positive note, nothing went bang.
Cheers
I stuck a couple of 5k6 in parallel on the front of my 160v regs. So
B+ supply (317v) ==> parallelled 5k6 5w ==> where it says B+ input on the schematic.
Thats working and showing 317v in and 158v out with about 115v drop accross the paralleled resistors (both 5w so essentially a 10w 2k8) which tallies with my calculations.
However My 260v reg is showing only 220v at its output. Pretty sure this was both sides so fairly sure its not the heater bias dragging it down.
Somethings getting hot, not sure yet what, small amount of smoke but cant see where its coming from and nothing appears to be burnt. It does seem to be coming from the 160v reg somewhere which is a bit odd as I would have thought that having put the extra resistors to drop the input voltage this would have helped.
Could it be that theres only about 40v buffer on that now? I think im getting 198v on the input proper of the 160v reg.
When i tested the psu on its own it didnt hum, but under load it is. Again not sure which tx this yet. Am a bit nervous of poking round too much. Also cant leave it on for very long to track down fault.
Would it be useful to remove the tubes for tracking this down?
The neons seem to be behaving and extinguishing after a few seconds.
I am wondering if my heater tx doesnt like having ~65 v on it but i think i have to have this to avoid damaging the tubes? Not only about reducing hum.
I think i need a clear strategy here to avoid breaking the whole thing or myself.
Would it help if i drew a diagram how ive wired it all up to make sure ive not done anything daft?
Sorry to be a bit vague, switch on time is somewhat limited until I decide on a plan.
On a positive note, nothing went bang.
Cheers
Hi mrdave,
Are you using one PS for both channels? You need two PS.
You can test step for step:
1. 260V testing without load. (I wrote, the values 68K an 220K from the circuit of SY give only 220V). This is a mistake. I used 77k and 270k.
2. If you have the 260V, simulate the load with resistor (13K).
3. 160V testing without load (use 52-53k and 165-170k).
4. Testing with load (3,2k).
Single testing both channels. if all okay test both channels.
Thats the way.
Send me your drawing to frankwilker@web.de.
Are you using one PS for both channels? You need two PS.
You can test step for step:
1. 260V testing without load. (I wrote, the values 68K an 220K from the circuit of SY give only 220V). This is a mistake. I used 77k and 270k.
2. If you have the 260V, simulate the load with resistor (13K).
3. 160V testing without load (use 52-53k and 165-170k).
4. Testing with load (3,2k).
Single testing both channels. if all okay test both channels.
Thats the way.
Send me your drawing to frankwilker@web.de.
I'm using 2 ht regs per channel. So I have 4 in total but that is powered from one raw psu. One 250-0-250 transformer supplies the lot.
Each reg should have 20ma draw on it.
I figured that each 5k6 resistor would see 20ma and so 2.2w per 5w device and so would be OK, but I guess that is quite a bit heat to lose.
I now see what you mean about the resistor values for setting the output voltage on the regs. I'll change the values for the 260v reg.
Presumably if I leave the 1k8 resistor in I shouldn't have to alter the other resistors on the 160v reg, especially if I am getting 158v anyway? Regardless of how it's arranged. If the input is 315v and the output is 160v, at 20ma draw then 3.1w has to be got rid of and this is just a question of how we optimally spread that across the various components.
Each reg should have 20ma draw on it.
I figured that each 5k6 resistor would see 20ma and so 2.2w per 5w device and so would be OK, but I guess that is quite a bit heat to lose.
I now see what you mean about the resistor values for setting the output voltage on the regs. I'll change the values for the 260v reg.
Presumably if I leave the 1k8 resistor in I shouldn't have to alter the other resistors on the 160v reg, especially if I am getting 158v anyway? Regardless of how it's arranged. If the input is 315v and the output is 160v, at 20ma draw then 3.1w has to be got rid of and this is just a question of how we optimally spread that across the various components.
Okay, now i´m understanding, how the 5k6 resistors are arranged. For durable using i would use a little more as 5W. I heared, a rule is, to use not more than a quarter of the load.
The 158V for the 160V regulator is okay. I think a little more input as 198V were better for filtering. But for testing it´s okay.
The 158V for the 160V regulator is okay. I think a little more input as 198V were better for filtering. But for testing it´s okay.
If the Voltage(s) are okay, wire only the 260V (and heater) to the HMN-Board. Then measure the voltage about R3 (1k). After a few seconds the voltage have to go to nearly 20V. Gratlations, the first stage is working.
Do the same with the second stage (only the second stage). The neon bulb will light on and after a few seconds go out. Instead the LED will slowly light on (not so bright).
Measure about R12 (1k) nearly 10V will be right. The first half of the second stage works.
Now measure the drop about your preresistor 5,6K. This drop minus 56V (for the 1st half/2nd stage for 10mA) gives the working current for the cathode follower.
You wrote the drop is 115V. This gives 59V/5,6k = voila nearly 10mA = perfect.
Now enable both stages and pull up the heater.
It´s weekend, so enjoy the sound.
Do the same with the second stage (only the second stage). The neon bulb will light on and after a few seconds go out. Instead the LED will slowly light on (not so bright).
Measure about R12 (1k) nearly 10V will be right. The first half of the second stage works.
Now measure the drop about your preresistor 5,6K. This drop minus 56V (for the 1st half/2nd stage for 10mA) gives the working current for the cathode follower.
You wrote the drop is 115V. This gives 59V/5,6k = voila nearly 10mA = perfect.
Now enable both stages and pull up the heater.
It´s weekend, so enjoy the sound.
heater cathode voltage
Thanks frank, ordered some more bits, will install them tomorrow when they arrive.
I dont think i understand this heater to cathode voltage correctly. Ive been looking into this due to wanting to test without heater elevation.
A few assumptions i want to make sure are correct.
Section 2 of the 6922 is the buffer (cathode follower) with the cathode at 95v.
Section 1 of the 6922 is the second gain stage with the cathode at 1.7v
Vkf is the notation for cathode heater (presumably f for filament)
In the article
"The more well-known twist is the use of R30 and R31 to elevate the heaters 65V above ground. This has two salutary effects - first, it reduces the heater-to-cathode strain of the cathode follower. For that tube, the cathode is roughly 95 volts above ground. Without this heater elevation, the heater-to-cathode voltage well exceeds the 50V limit for triode section one and strains the limits of triode section two. With 65 volts of elevation, the heater-cathode potential for the follower is a balmy 30V. "
I looked up the datasheet for the 6922eh. It seems that this can have the heater to cathode positive or negative 200v. So for this tube i dont think this applies.
I looked up the ecc88 and found a philips ecc88. Im assuming this is a similar tube to the one sy uses. This has the following
Vkf = 50v max
Vk'f (k' pos; f neg) = 150v max
I have also seen from telefunken CCa
Uf/k+ = 150v
Uf/k- =100v
The way the article is worded it looks at if Vkf is when the cathode is positive relative to the heater and Vk'f is when the cathode is negative relative to the heater.
I would have thought that Vk'f (k' pos; f neg) = 150v means that the cathode can be a maximum of 150v higher than the heater.
Even so, if the heater cathode limit for the follower is 50v and thats how to read the spec sheet, i can see that the 95v exceeds this and so the heater elevation fixes it but why does this strain the other section without the heater elevation? Surely the cathode for the second gain stage would be at ~1.7 volts and the heater at somewhere between -6.3v to zero or zero to +6.3v. so only a few volts difference, well within any of the specs.
Ok, going on the siemens D3a. The spec sheet i found for that has this.
Voltage between cathode and heater
Cathode positve Vkf = max 120v
cathode negative Vkf = max 60v
To me this is implying that if I elevate the heater to +65 volts then the cathode is going to be about 64v below the heater and is likely to be damaged.
I've catastrophically misunderstood something here but Im not sure where ive gone wrong.
Thanks frank, ordered some more bits, will install them tomorrow when they arrive.
I dont think i understand this heater to cathode voltage correctly. Ive been looking into this due to wanting to test without heater elevation.
A few assumptions i want to make sure are correct.
Section 2 of the 6922 is the buffer (cathode follower) with the cathode at 95v.
Section 1 of the 6922 is the second gain stage with the cathode at 1.7v
Vkf is the notation for cathode heater (presumably f for filament)
In the article
"The more well-known twist is the use of R30 and R31 to elevate the heaters 65V above ground. This has two salutary effects - first, it reduces the heater-to-cathode strain of the cathode follower. For that tube, the cathode is roughly 95 volts above ground. Without this heater elevation, the heater-to-cathode voltage well exceeds the 50V limit for triode section one and strains the limits of triode section two. With 65 volts of elevation, the heater-cathode potential for the follower is a balmy 30V. "
I looked up the datasheet for the 6922eh. It seems that this can have the heater to cathode positive or negative 200v. So for this tube i dont think this applies.
I looked up the ecc88 and found a philips ecc88. Im assuming this is a similar tube to the one sy uses. This has the following
Vkf = 50v max
Vk'f (k' pos; f neg) = 150v max
I have also seen from telefunken CCa
Uf/k+ = 150v
Uf/k- =100v
The way the article is worded it looks at if Vkf is when the cathode is positive relative to the heater and Vk'f is when the cathode is negative relative to the heater.
I would have thought that Vk'f (k' pos; f neg) = 150v means that the cathode can be a maximum of 150v higher than the heater.
Even so, if the heater cathode limit for the follower is 50v and thats how to read the spec sheet, i can see that the 95v exceeds this and so the heater elevation fixes it but why does this strain the other section without the heater elevation? Surely the cathode for the second gain stage would be at ~1.7 volts and the heater at somewhere between -6.3v to zero or zero to +6.3v. so only a few volts difference, well within any of the specs.
Ok, going on the siemens D3a. The spec sheet i found for that has this.
Voltage between cathode and heater
Cathode positve Vkf = max 120v
cathode negative Vkf = max 60v
To me this is implying that if I elevate the heater to +65 volts then the cathode is going to be about 64v below the heater and is likely to be damaged.
I've catastrophically misunderstood something here but Im not sure where ive gone wrong.
You are right, the cathode can be the Ufk higher than the heater. Otherwise the isolation between heater and cathode can be damaged or affects a hum or other noises.
All you wrote is right, but you measuring is false. The voltage about Rk is 1,7V. Now measure from Ua to k and you found nearly 140V? So Uk is Ua - Urk (I hope all notations are right). It´s only a thinking fault.
So the voltage is above the Ufk. To heighten the Heater gives a smaller difference. But attention, the heater is not connected to Common. Only about the voltage dividier. (C is much to great. Enough is up to 10 µf.)
A cathode follover is in the most cases no problem. The voltage is a little higher than the half of Va, here like 90V, the heater is 65V, so the difference is 25V.
I hope someone understand my bad english, sorry.
All you wrote is right, but you measuring is false. The voltage about Rk is 1,7V. Now measure from Ua to k and you found nearly 140V? So Uk is Ua - Urk (I hope all notations are right). It´s only a thinking fault.
So the voltage is above the Ufk. To heighten the Heater gives a smaller difference. But attention, the heater is not connected to Common. Only about the voltage dividier. (C is much to great. Enough is up to 10 µf.)
A cathode follover is in the most cases no problem. The voltage is a little higher than the half of Va, here like 90V, the heater is 65V, so the difference is 25V.
I hope someone understand my bad english, sorry.