N1ESE said:
N1ESE said:
Ok, something is not adding up (but we knew that already). With 1v (close enough) on the grid and 3.2v on the plate of the driver, you are getting a gain for this stage of, well, 3.2. That's way too low. With an AT7 and a bypassed cathode resistor, I'd expect a gain of at least 30.
Second problem is that the output at the plate of the T7 should be the same as the grid of the output tube.
Are these measurements all taken with the same input to the amp?
The gain at the output tube is about 12. That's at least in the right ballpark. A voltage ratio of 24:1 from plate to output would be about right for a 4.5k OPT.
So, let's take a closer look at that input stage or between the input and output.
Sheldon
Ok, I've removed the pot from the circuit. Computer soundcard is now hardwired directly to the amplifier. Still generating the same 1 KHz tone at roughly 1 VAC.
I've started by taking a closer look at the input tube plate voltages and something is troubling me. The channels aren't balanced, I am seeing different voltages on each side.
Left Channel:
3.8 VAC with tone, 5.8 VAC with no tone
Right Channel:
4.8 VAC with tone, 7.0 VAC with no tone
Both readings were taken just before the CCS plate resistors (R13/R23).
Edit: Nate, thanks!
I've started by taking a closer look at the input tube plate voltages and something is troubling me. The channels aren't balanced, I am seeing different voltages on each side.
Left Channel:
3.8 VAC with tone, 5.8 VAC with no tone
Right Channel:
4.8 VAC with tone, 7.0 VAC with no tone
Both readings were taken just before the CCS plate resistors (R13/R23).
Edit: Nate, thanks!
With no input signal applied, there shouldn't be any AC signal at the plate of the 12AT7. Most of the evidence seems to suggest that 12AT7 is a dud. Do you have a pellet gun? Vacuum tubes make for excellent target practice.
I wonder if your meter gives reliable AC readings when there is a substantial amount of DC voltage present. Maybe that's why the "math doesn't add up" when you're checking the AC voltages on either side of the C11 coupling cap.
Sheldon said:
I wonder if your meter gives reliable AC readings when there is a substantial amount of DC voltage present. Maybe that's why the "math doesn't add up" when you're checking the AC voltages on either side of the C11 coupling cap.
N1ESE said:
I agree with Ty on this one. There should be very little AC on the plate with no input input signal. I assume you are measuring everything with respect to signal ground. If you removed the pot, you do have the 220k to ground still? What is the DC voltage at the plate again? Also, just to check, try measuring with a cap (0.1uF, or thereabouts, film) in series with the + multimeter probe. If the measurements are different than those without, use the cap.
The fact that you get different measurements on either side of the interstage cap, makes me wonder if your multi meter is thrown off by the DC offset, as the input plate will be in the 100V range, for example (I don't know the specific figure for this amp), while the output grid will be at ground potential. Make sure the measurement system is sorted out first.
Sheldon
You know what guys, I'm going to give up on this project and sell the components. It's just not working out like I thought and I've had some other things come up. I thank everyone for their time and help.
Nate, I'm going to send the tubes back to you, with some cash for your shipping expense, when they arrive. Thank you very much for this.
I will be posting the iron and stuff for sale here on the Trading Post shortly. If any of you want the board assembly, I'm going to be selling this very cheaply. PM me if anyone wants any of this.
I'm not giving up completely. I'm probably going to build a Millet MOSFET-MAX headphone amp next. Probably after the holidays.
Nate, I'm going to send the tubes back to you, with some cash for your shipping expense, when they arrive. Thank you very much for this.
I will be posting the iron and stuff for sale here on the Trading Post shortly. If any of you want the board assembly, I'm going to be selling this very cheaply. PM me if anyone wants any of this.
I'm not giving up completely. I'm probably going to build a Millet MOSFET-MAX headphone amp next. Probably after the holidays.
I am fine, after 6 surgeries I think they got it right! The corporate hatchet swingers have been laying people off again so I got to stay off the forums from work for the forseable future. Unfortunately I have been living there lately. Last time I checked in you were still waiting on transformers. Board sales have been very slow, so I spent my free time over the past weekend catching up on what has happened in the last 20 years in the ham radio world. My last radio had tubes in it. Now I have a tiny Yaesu that covers 160m to 432, and I am scratching my head over JT65.
I looked over your pictures and everything is wired up correctly. It seems that the consensus is that you have low gain in the input stage. I tend to agree. The first suspect is the 12AT7 tube. It may not even be a 12AT7, since you say that it isn't marked. Try a replacement when it arrives. The DC voltage readings that you provided look OK which indicate that the input tube is at least baising up somewhat like a 12AT7. A 12AU7 or other 12A*7 tube can be made to work but usually requires changing some resistor values.
As stated before you should have 10 to 20 volts AC on the output tube grid. This should be measured where C11 connects to R15 and R16. Most meters will not read AC in the presence of a DC voltage as is the case on the other side of C11.
The 12AT7 in the Simple SE is CCS loaded so the gain is higher, I see 45 to 55 depending on the tube. Maybe there is another cause for the low gain. I can't quite make out the colors on the resistors in your photo. Are R16 and R26 the correct value? Anything from 100K to 270K should work, but if the resistor value is too low the gain will be reduced. Check them with an ohm meter to be sure. I have seen mismarked resistors before (only once though). The resistors are about the only thing that I can't see in your photo, so it wouldn't hurt to check them over with a meter while you are waiting for a tube.
With no input signal, verify that about 100 volts DC is being dropped across R14 (measure each end and subtract). This verifies that the 12AT7 is drawing about 10 mA and the CCS is working correctly.
DON'T GIVE UP!
You are so very close! You have some great people here trying to help you. Remember, you learn the most from your mistakes! I would see if the tube is a dud, and then talk to those here trying to help. Remember also that others will read this and add to the collective knowledge base.
Hang in there! This amp is great, I have built half a dozen or so tube amps now, the Simple SE was the first, and it is still my main amp!
You now have Tubelab here helping, I am sure that it will be sorted out very soon.
It will be worth the hard work!
Chris
You are so very close! You have some great people here trying to help you. Remember, you learn the most from your mistakes! I would see if the tube is a dud, and then talk to those here trying to help. Remember also that others will read this and add to the collective knowledge base.
Hang in there! This amp is great, I have built half a dozen or so tube amps now, the Simple SE was the first, and it is still my main amp!
You now have Tubelab here helping, I am sure that it will be sorted out very soon.
It will be worth the hard work!
Chris
tubelab.com said:
460 on one side and 380 on the other so about 80 volts.
Resistor values are one of the first things we checked.
Glad you are in good health and surviving the hatchet men.
I've been over this thing with a fine tooth comb like crazy the past few days and I just found out I have to have oral surgery next week that I have to pay for out of pocket. Quite frustrating to have ~$700 in parts sitting here not working properly right now.
N1ESE, I was writing while you were posting. Don't give up yet. If you can't figure it out, email me and I will make arangements to get the board working. I have been selling tube amp boards for 3 years. So far there have been only two cases (both Tubelab SE's) where I had the builders send me their boards and I made them work. Usually they can be made to work remotely. With my crazy schedule it may take a few days, but I will make it work.
The Simple SE will make plenty of sound with your speakers. I am using inefficient (87db) Yamaha's and Sherri is always telling me to turn it down.
I understand your frustration, but it will be worth it when you have it working. I am now on my third attempt at a simple 20 meter dipole and I am an RF engineer.
The Simple SE will make plenty of sound with your speakers. I am using inefficient (87db) Yamaha's and Sherri is always telling me to turn it down.
I understand your frustration, but it will be worth it when you have it working. I am now on my third attempt at a simple 20 meter dipole and I am an RF engineer.
tubelab.com said:
73 DE Z37HWX!
Nice to see we are all around.
Use lambda/2*shorting factor (depends of diameter of wire or tube)
For 20 meters you should have little bit short than 10 meters for total length of the dipole.
Than use SWR meter and do a fine adjust to the lenght to get minimal SWR @ the working frequency.
I have measured out 10 meters of wire, but it is likely being detuned by proximity to the roof. I borrowed the radio last week and just decided to "throw a wire into the air" so that I could RX some JT65 signals on 20 to verify the computer setup. I am designing an SDR similar to a Flex Radio 5000. Long term plans are for EME work.
As N1ESE is probably frustrated by trying to troubleshoot an amplifier without a scope, I have a similar issue. I have no means of measuring VSWR. I can only guess, key the radio and watch the HSWR icon flash on the display. I sold all of my ham equipment about 15 years ago when the local city code enforcers banished my antennas, hence the dipole a few inches above the roof line.
Close enough. The absolute accuracy of the CCS IC's is not that great.
That would be my first guess. The Russians are rather liberal in their use of US type numbers, and some Ebay sellers will take evem more liberty with them. I have seen strange things happen if you use a 12AX7 tube in a Simple SE since some of them really don't like 10 mA (or 8 mA).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.