Hi guys,
I just finished building a Elekit TU-8900 SE amp with a pair of WE300B and encountered an issue during initial tests.
The Elekit amp uses a cathode bias and is running with these measured parameters:
Ub = 390V
Uk = 62V
Rk = 810 Ohm
=> Ik = 62V / 810 Ohm = 76mA
I verified the Ik of 76mA by measuring a voltage drop of 8V and a DCR of 106 Ohm over the output transformer (Lundahl LL2785C).
8/106 = 76mA
So, plate current seems to be fine either.
Then I measured
Ug Bias voltage (cathode to grid): -62V
The grid leakage resistor Rg is 250k (220k + 15k +15k), which is the maximum value for self bias according to the WE300B datasheet.
So far, nothing fancy, nothing special, I guess 🙂
But and here is my issue, Elekit list some voltages to be checked in their manual to approve a proper build and operation.
According to the Elekit manual, the voltage between Grid and Gnd (AKA DC offset at grid) should be measured with "0V (< 30mV)" in Idle mode after some heating up time.
I don't understand this value 🤔
Assuming no other current coming from a leaking coupling cap, the only current, which contributes to the DC offset is the Grid current.
To measure a value below the given 30mV, the Grid current must be smaller than:
Ig < 30mV / 250k = 0,12uA
In the WE300B datasheet a max. Grid current of 1uA is specified.
In the Test Certificate of my pair of WE300B the Grid currents are: 0.86uA and 1.00uA
Using these values will result in a DC offset at the Grid of:
0.86uA * 250k = 215mV
1uA * 250k = 250mV
So, more than 7 times higher than specified by Elekit! 🤔
I measured the tubes with my Tube tester and they are absolutely fine.
The Tube tester measured grid current is 0,389uA and 0.291uA, which result in
0.389 * 250k = 97mV
0.291 * 250k = 73mV
And these are the values, which I measure in the amp.
Still approx. 3 times higher than the 30mV given by Elekit....
I have a set of Chinese 300B here, which tested fine on my Tube tester either.
The results are compareable with the WE300B.
The grid currents for both tubes are well below 1uA (0,3uA and 0,2uA) and will result in a measured DC offset of 75mV and 50mV.
Why is Elekit specifying such a small value of an "allowed" DC offset below 30mV, which will result in a maximum grid current of 0,12uA?
Keeping in mind, that even WE "allows" 1uA in their High-End and High-$$$ tubes, which will result in a DC offset of approx. 250mV based on a Rg of 250k.
I doubt, that 0,12uA is a value reached by almost all 300Bs out there and that I have just got 4 bad tubes here.
Is a Grid DC offset of 250mV an issue of concern?
Am I doing something wrong or misunderstand here?
I'm puzzled 🙂
Jochen
I just finished building a Elekit TU-8900 SE amp with a pair of WE300B and encountered an issue during initial tests.
The Elekit amp uses a cathode bias and is running with these measured parameters:
Ub = 390V
Uk = 62V
Rk = 810 Ohm
=> Ik = 62V / 810 Ohm = 76mA
I verified the Ik of 76mA by measuring a voltage drop of 8V and a DCR of 106 Ohm over the output transformer (Lundahl LL2785C).
8/106 = 76mA
So, plate current seems to be fine either.
Then I measured
Ug Bias voltage (cathode to grid): -62V
The grid leakage resistor Rg is 250k (220k + 15k +15k), which is the maximum value for self bias according to the WE300B datasheet.
So far, nothing fancy, nothing special, I guess 🙂
But and here is my issue, Elekit list some voltages to be checked in their manual to approve a proper build and operation.
According to the Elekit manual, the voltage between Grid and Gnd (AKA DC offset at grid) should be measured with "0V (< 30mV)" in Idle mode after some heating up time.
I don't understand this value 🤔
Assuming no other current coming from a leaking coupling cap, the only current, which contributes to the DC offset is the Grid current.
To measure a value below the given 30mV, the Grid current must be smaller than:
Ig < 30mV / 250k = 0,12uA
In the WE300B datasheet a max. Grid current of 1uA is specified.
In the Test Certificate of my pair of WE300B the Grid currents are: 0.86uA and 1.00uA
Using these values will result in a DC offset at the Grid of:
0.86uA * 250k = 215mV
1uA * 250k = 250mV
So, more than 7 times higher than specified by Elekit! 🤔
I measured the tubes with my Tube tester and they are absolutely fine.
The Tube tester measured grid current is 0,389uA and 0.291uA, which result in
0.389 * 250k = 97mV
0.291 * 250k = 73mV
And these are the values, which I measure in the amp.
Still approx. 3 times higher than the 30mV given by Elekit....
I have a set of Chinese 300B here, which tested fine on my Tube tester either.
The results are compareable with the WE300B.
The grid currents for both tubes are well below 1uA (0,3uA and 0,2uA) and will result in a measured DC offset of 75mV and 50mV.
Why is Elekit specifying such a small value of an "allowed" DC offset below 30mV, which will result in a maximum grid current of 0,12uA?
Keeping in mind, that even WE "allows" 1uA in their High-End and High-$$$ tubes, which will result in a DC offset of approx. 250mV based on a Rg of 250k.
I doubt, that 0,12uA is a value reached by almost all 300Bs out there and that I have just got 4 bad tubes here.
Is a Grid DC offset of 250mV an issue of concern?
Am I doing something wrong or misunderstand here?
I'm puzzled 🙂
Jochen
1µA limit for power-valve grid leakage is normal; and the 300Bs are running very cool in this operating point. You can usually scale the grid leak resistor upward, in proportion to the anode power burn decrease from the maiximum allowed [36W for most 300Bs]
So I suspect a typo in the amplifier manual. Write to the manufacturer for confirmation.
If you prefer the tightest control on the DC anode current, decrease the grid leak R value [and increase the coupling cap in proportion].
So I suspect a typo in the amplifier manual. Write to the manufacturer for confirmation.
If you prefer the tightest control on the DC anode current, decrease the grid leak R value [and increase the coupling cap in proportion].
@Rod Coleman
Thanks for your reply! I will encounter your proposal.
@vkung
I bought a pair of WE300B in the Wood box from BTB Elektronik here in Germany. BTB is a official Western distributor and probably one of the largest and well known tube sellers here in Germany. This is definitely no B Stock.
https://btb-elektronik.de/we300b-we...factory-matched-pair-holz-box/300b-west.1paar
As I wrote: 1uA is given in the WE300B Datasheet and values of 0.86uA and 1.00uA for Grid Current (Ic) are printed on the Test Certificate of the two tubes.
All other voltage measurements are fine. I will send the list later....I'm not in my workshop right at the time
@waltube
I'm using a RoeTest V11. Probably the king of Tube Testers ;-)
This is my device:
https://www.roehrentest.de/NB Hammer Jochen.html
Thanks for your reply! I will encounter your proposal.
@vkung
I bought a pair of WE300B in the Wood box from BTB Elektronik here in Germany. BTB is a official Western distributor and probably one of the largest and well known tube sellers here in Germany. This is definitely no B Stock.
https://btb-elektronik.de/we300b-we...factory-matched-pair-holz-box/300b-west.1paar
As I wrote: 1uA is given in the WE300B Datasheet and values of 0.86uA and 1.00uA for Grid Current (Ic) are printed on the Test Certificate of the two tubes.
All other voltage measurements are fine. I will send the list later....I'm not in my workshop right at the time
@waltube
I'm using a RoeTest V11. Probably the king of Tube Testers ;-)
This is my device:
https://www.roehrentest.de/NB Hammer Jochen.html
I want to see your test points.
JHammer
I assembled more than 50`~60 units and fix several TU-8900 in the last 2.5 yeas. The majority issues are the soldering and bad tube.
@vkung
Thank you very much for your quick reply. I really appreciate your support!
I have more than 30 years experience in developing (audio) hardware and developed and build several tube amps in the past.
This is the reason, why I own a High-End Tube Tester and other audio test gear like an Audio Precision Analyzer.
This is my first SE amp with 300B, however.
I triple checked all soldering and measured all values of all the parts several times.
One test, for example, was to lift the coupling caps to see if they might leak.
Then I removed C107/207 to see, if this changed something. Unfortunately not. The measured DC offset is still present and depends on the tube I'm using (out of the 4 I have here).
Hmm...hard to believe.
I'd like to understand, why Elekit specifies a grid current of 0.12uA, which is required for an offset below 30mV with the Rp of 250k in total,
whereas Western (and all other sources in the web) states, that a value below 1uA is okay.
Isn't the Grid DC offset caused by the Grid Current, which flows from Grid to GND through the Grid leakage resistor?
We have a 250k resistor (220k + 15k +15k). So by a Grid current of 1uA, this results in 250mV voltage over the resistors.
Do I miss an important point here ?
I will supply the test points asap when I'm back in my workshop...
Thank you very much for your quick reply. I really appreciate your support!
I'm aware of this and read all posts about the TU-8900 in this and other forums several times in order to understand what could cause my issue...
I have more than 30 years experience in developing (audio) hardware and developed and build several tube amps in the past.
This is the reason, why I own a High-End Tube Tester and other audio test gear like an Audio Precision Analyzer.
This is my first SE amp with 300B, however.
possible, but when I swap the tubes, the measured DC offset swaps either. And it's present on both channels.
I triple checked all soldering and measured all values of all the parts several times.
One test, for example, was to lift the coupling caps to see if they might leak.
Then I removed C107/207 to see, if this changed something. Unfortunately not. The measured DC offset is still present and depends on the tube I'm using (out of the 4 I have here).
All 4 of them brand new out of the box and tested good in my Tube Tester?
Hmm...hard to believe.
I'd like to understand, why Elekit specifies a grid current of 0.12uA, which is required for an offset below 30mV with the Rp of 250k in total,
whereas Western (and all other sources in the web) states, that a value below 1uA is okay.
Isn't the Grid DC offset caused by the Grid Current, which flows from Grid to GND through the Grid leakage resistor?
We have a 250k resistor (220k + 15k +15k). So by a Grid current of 1uA, this results in 250mV voltage over the resistors.
Do I miss an important point here ?
I will supply the test points asap when I'm back in my workshop...
I always ask customer to assemble according to the manual before you change the original design
I assume you did but follow the manual to build the amp
why you put two coupling caps in the slot? It is dangerous to install the resistors in this manner
I need your test points to assist you
I assume you did but follow the manual to build the amp
why you put two coupling caps in the slot? It is dangerous to install the resistors in this manner
I need your test points to assist you
I followed the manual to build the amp and have not changed the design and any of the part values. I replaced the kit resistors with resistors from Audio Note, AMRG, Mills and caps from Audio Note, Mundorf and Panasonic 35SEK330M instead of the Kemets, because they have a lower ESR.
I added an additional 10nF Silver foil Cap from Duelund in paralled to the 100nF Copper foil from Jupiter as the coupling caps.
But, I already lifted them on the grid side to check for leakage. No change in the DC Grid offset.
Here are the tespoints from a measurement I did right now.
AC: 224V AC here in my workshop at the time of measurements
Last night, I had 234V AC and all relevant values 7, 8, 15...22 were approx. 5 to 10V higher
1./2. 2.64/2.66
3./4. 60/57
5./6. 66/64
7./8. 200/203
9./10. 0,13V / 0,21V <= noisy and increases slightly => thermal drift?
11./12. 59/60
13./14. 64/65
15./16. 370/370
17./18. 378/379
19./20. 299/299
21./22. 399/398
23. 5.6
I measured the overall cathode resistor
R117 + R118 + R199: 809 Ohm
With TP11: 59V => 73mA for Ik
R217 + R218 + R299: 810 Ohm
With TP12: 60V => 74mA for Ik
I checked the voltage drop on the Lundahls and measured its DRCs
7.98V @ 107,8 Ohm => 74mA for Ip
8.04V @ 108,0 Ohm => 74,4mA for Ip
I'm really curious what happens here....
I added an additional 10nF Silver foil Cap from Duelund in paralled to the 100nF Copper foil from Jupiter as the coupling caps.
But, I already lifted them on the grid side to check for leakage. No change in the DC Grid offset.
Here are the tespoints from a measurement I did right now.
AC: 224V AC here in my workshop at the time of measurements
Last night, I had 234V AC and all relevant values 7, 8, 15...22 were approx. 5 to 10V higher
1./2. 2.64/2.66
3./4. 60/57
5./6. 66/64
7./8. 200/203
9./10. 0,13V / 0,21V <= noisy and increases slightly => thermal drift?
11./12. 59/60
13./14. 64/65
15./16. 370/370
17./18. 378/379
19./20. 299/299
21./22. 399/398
23. 5.6
I measured the overall cathode resistor
R117 + R118 + R199: 809 Ohm
With TP11: 59V => 73mA for Ik
R217 + R218 + R299: 810 Ohm
With TP12: 60V => 74mA for Ik
I checked the voltage drop on the Lundahls and measured its DRCs
7.98V @ 107,8 Ohm => 74mA for Ip
8.04V @ 108,0 Ohm => 74,4mA for Ip
I'm really curious what happens here....
12BH7 yes
Sorry, I forgot to add that information
Both LEDs change to blue and the two orange LEDs in each of the modules light up either.
This measurement was taken with the Chinese 300B, because I want to avoid damaging the $$$ WEs
I replaced them with the WE300B already two days ago with no change in the result.
So, I changed back to the Chinese one.
Sorry, I forgot to add that information
Both LEDs change to blue and the two orange LEDs in each of the modules light up either.
This measurement was taken with the Chinese 300B, because I want to avoid damaging the $$$ WEs
I replaced them with the WE300B already two days ago with no change in the result.
So, I changed back to the Chinese one.
try to replace the driver to 12AU7 and measure the test points..
you reading is lower than the ref test points.
you should have sound is it right?
you reading is lower than the ref test points.
you should have sound is it right?
Sound is fine from listening point of view. Nothing to complain, no hum, no noise...
However, I haven't checked with my Audio Analyzer so far.
12AU7...must have a look at my stock if I have some available...
I already made some tests and unplugged the 12BH7s and just ran the 300 tubes. No difference regarding the offset on TP9/10
However, I haven't checked with my Audio Analyzer so far.
12AU7...must have a look at my stock if I have some available...
I already made some tests and unplugged the 12BH7s and just ran the 300 tubes. No difference regarding the offset on TP9/10
from your test points nothing to worry
your amp is running ok.. but the voltage is lower than the average
what is your power main input?
TP9/TP10 should be very low.. that related to your coupling cap and 300B
If you roll back to stock 0.1 cap your PT 9 and PT10 should be lower
your amp is running ok.. but the voltage is lower than the average
what is your power main input?
TP9/TP10 should be very low.. that related to your coupling cap and 300B
If you roll back to stock 0.1 cap your PT 9 and PT10 should be lower
AC: 224V AC here in my workshop at the time of measurements
Last night, I had 234V AC and all relevant values 7, 8, 15...22 were approx. 5 to 10V higher
But, as I wrote:
I already lifted them on the grid side to check for leakage. No change in the DC Grid offset at TP9/10
Last night, I had 234V AC and all relevant values 7, 8, 15...22 were approx. 5 to 10V higher
Ok, I will try
But, as I wrote:
I already lifted them on the grid side to check for leakage. No change in the DC Grid offset at TP9/10
These curves look a bit funky to me. The tube does not want to turn off from 150 V and up.