My two kits finally arrived today! Box indicates that they were sent from California. Each kit was individually sealed in its own anti-static bag with the tubes bubble wrapped for protection. The 6J1 tubes are really small, about half the size of 12AX7s. I am tempted to just quickly populate my boards and fire them up. But I think I will resist and start by doing an inventory all parts, measuring and logging each one. I'd also like to test the 6J1 tubes on my tube tester, but I don't think I have this tube type on any of my roll charts. Maybe the GE5654 tube type is listed. I will have to check.
I'm curious if you are missing any parts. Probably not, seeing how well your are packaged. I actually got an extra diode in mine! Since you have the power supplies already breadboarded, I'll bet you'll be up and running in no time!
I dont see that tube listed either on the heathkit roll chart on mine. Maybe came out in later years? I wonder if there is another tube that can be cross referenced ?
I ran across this alternate circuit that uses a single 240v supply. I may try that with my second kit. That also gives me an excuse to use my old HV tube power supply.
I ran across this alternate circuit that uses a single 240v supply. I may try that with my second kit. That also gives me an excuse to use my old HV tube power supply.
The 6AK5 is on my tube tester charts. Can't find the 5654 on my charts, so that tube must have been released after my charts were printed. I'll do some more on-line checking for updated tube charts for my tester. I also need to double check and make sure these other tubes are pin-for-pin drop in mechanical and electrical replacements for the 6J1. Sometimes a tube might be listed as a substitute, but could have different electrical characteristics or pinout connections.
@Tom_S said:
I'm curious if you are missing any parts. Probably not, seeing how well your are packaged. I actually got an extra diode in mine! Since you have the power supplies already breadboarded, I'll bet you'll be up and running in no time!
I've only opened and tested one kit so far. No missing parts. I tested all the resistors and they all appear to be very tight tolerance, 1 to 2% of nominal value. All 8 of the 470uF caps measure about 445uF, about 5% low. They sent four 1uF ceramic disc 50V coupling caps. They test OK, but I am going to replace them with some small 64V polyester caps that I have on hand. I think you replaced your 1uF coupling caps as well. All transistors and diodes check OK. The 50K pot measures 50.8K and seems to track well channel to channel. So far, so good.
Tom, I see you substituted Nichicon ES series 3.3uf bi-polar caps for the input and output coupling caps. Looks like a really tight fit next to the resistors. I found some WIMA 3.3uF 50V polyester caps in my unused parts box. They are about the same size as the Nichicon ES 3.3uF's, but they have very short leads. So I will have to solder extensions onto the existing leads so that I can mount them slightly proud of the resistors underneath them.
Bill, the 5654 is a ruggedized version of the 6AK5, from what I have read. I have them running in mine now. I'll have to double check, but I think the distortion is a little higher with the 5654 tubes. But I don't hear any sound when I tap them like I did with the stock tubes.
Glad you weren't missing any parts! I know there had been some complaints about missing parts on some of the Amazon listings, but there seem to be a few different sellers of the same basic kits. I was going to put some Wima or Kemet film caps in place of those ceramic couplers, but like you said, it's pretty tight between the resistors. I have a ton of the ES caps, so I thought I would use those instead. They sound and measure fine to me.
I was curious, so I hooked up the heaters directly to the little 12.3vac transformer. There is some hum, but it's pretty low when listening with headphones. I'm just guessing, but I think the AC is being picked up on adjacent traces somewhere. I'll still end up using a diode, cap, and the little regulator board.
@4thtry said:
They sent four 1uF ceramic disc 50V coupling caps. They test OK, but I am going to replace them with some small 64V polyester caps that I have on hand.
In college I built a guitar sustain unit as a final project. I put a ceramic cap in the 1st stage and it worked fine on the bench but if you tried to use in the real world it would scream with feedback. Ceramic disks can be very microphonic and I would never use one in an audio circuit.
I tested my set of four 6J1 tubes on my Heathkit TT-1 tube tester, which can measure transconductance (GM) in micromhos. I used the 6AK5 roll chart data to set up the tester, so this may not be entirely accurate for a 6J1 tube.
So, it looks like I lucked out by ordering two kits. I can get a good match by using the 3550/3450 pair.
EDIT: I just measured the heater voltages in the final series circuit and the way the voltage splits between the two tubes appears to be closely related to the measured GM value. When I plug the matched set of 3550/3450 tubes into the sockets, I get the same heater voltage at pins 3 & 4 of each tube: 6.15V. But when I plug the unmatched set of 2800/4500 tubes into the sockets, I get a heater voltage of 5.8 volts for the 2800 tube and 6.25 volts for the 4500 tube. This makes perfect sense, because transconductance goes up when the filament burns hotter.
Wow - the pairs packaged with each kit are pretty far apart. Glad yo got a decent pair to work with. I'm not sure how much variance is enough to cause an imbalance in this design, but now I'll be testing mine too.
I finished populating my main signal path PCB. I added some jumpers to convert the board from a dual polarity +28, 0, -28 volt supply to a single polarity +58, 0 volt supply. Instead of using my second PCB for the other power supply parts, I decided to make a small custom circuit board from scratch. This was much easier than modifing my second PCB because that would have required cutting several circuit board traces and I didn't want to do that. So now all I have to do is install the two boards in my main cabinet and power it up. I put the rectifiers and first RC smoothing stages next to the transformers.
I used a small PCB prototyping board to lay out the hole spacing and then taped this prototyping board to a single sided board. Next, I drilled all the holes with a #60 drill bit using my drill press. I then buffed the copper to a high sheen with #0000 steel wool and cleaned the surface with rubbing alcohol. Next, I painted all the traces free hand using a king size Sharpie. After gloving up, I filled a small plastic tray with a small amount of ferric chloride and set the board in the tray, copper side up. Using a small, wooden stick, I lightly agitated the board back and forth until all the copper was removed between the traces. This took about 20 minutes. To stop the etching process, I rinsed the board in running water.
I then cleaned the Sharpie ink from the traces with rubbing alcohol and disposable paper towels. To keep the copper from oxidizing, I tinned the traces with rosen flux and a small amount of solder. You can also tin the traces with a tinning solution, but I didn't have any tinning solution on hand.
I built a few pcb's back when Radio Shack had those kits. Also had a construction class in college and we had to design and make our own pcbs. Funny, I really loved that class
I was once set up to screen print PC boards and etch them. I got tired of drilling all the holes so I would send my artwork off to an outfit that would make the boards for me. I had a cad program that was dedicated to PCB layouts and once did a custom layout for a audio outfit here in Iowa. I still have equipment with some of the boards I made.
I used to roll my own preamps, power supplies and active crossovers.
I have all the parts for the chassis, I think. But I need to put my project on hold for a bit. I have a vintage receiver restoration for a paying customer on the bench.
Woohoo! Success! Powered it up with dim bulb and variac late last night. No smoke or fuses blown, so I gave it full power. My targeted B+ of 56VDC actually came out to 58.3VDC (close enough). My targeted heater voltage of 6.3VDC actually came out to about 6.15VDC on each tube (close enough). Ripple on the B+ and heater supply was unmeasureable on my scope, probably somewhere in the 2 to 4mV rms range. There was no hum at all from my bench speakers, even with the volume turned all the way up and my ear next to the woofer cones. I connected the outputs to my small Lepai 2020, which has an input impedance of 10.9K ohms. Driving into this load, preamp gain at 1kHz was 11.19dB in one channel and 10.88dB in the other channel. I played several music selections from my old cell phone and it really sounded good. So far, so good. Now I can put the front panel and top cover in place and do some frequency response measurements. This preamp has a very high output impedance, so I want to test to see how badly the high frequencies roll off when driving into a 3 to 6 foot long RCA cable and connected to a 10k ohm load.
Looking good, Bill! Your power supply is probably the biggest ever used with this little kit. I'm curious to see your measurements. Have you done the feedback and other mods yet? I see the Wima caps.
Comments
My two kits finally arrived today! Box indicates that they were sent from California. Each kit was individually sealed in its own anti-static bag with the tubes bubble wrapped for protection. The 6J1 tubes are really small, about half the size of 12AX7s. I am tempted to just quickly populate my boards and fire them up. But I think I will resist and start by doing an inventory all parts, measuring and logging each one. I'd also like to test the 6J1 tubes on my tube tester, but I don't think I have this tube type on any of my roll charts. Maybe the GE5654 tube type is listed. I will have to check.
The 6J1 tubes are 1/2 the size of a 12AX7? That is tiny!
There are lots of 12AX7 tubes to roll with, and consider upgrading a few caps… have fun
I'm curious if you are missing any parts. Probably not, seeing how well your are packaged. I actually got an extra diode in mine! Since you have the power supplies already breadboarded, I'll bet you'll be up and running in no time!
I dont see that tube listed either on the heathkit roll chart on mine. Maybe came out in later years? I wonder if there is another tube that can be cross referenced ?
The GE 5654 is the American equivalent of the Asian 6j1.
The 6AK5 is listed as an equivalent.
I ran across this alternate circuit that uses a single 240v supply. I may try that with my second kit. That also gives me an excuse to use my old HV tube power supply.
Scoll down -
https://www.diyaudio.com/community/threads/fx-audio-tube-01-mods.354479/
Here is a side by side comparison of a GE 12AX7 next to the 6J1 tube.
The 6AK5 is on my tube tester charts. Can't find the 5654 on my charts, so that tube must have been released after my charts were printed. I'll do some more on-line checking for updated tube charts for my tester. I also need to double check and make sure these other tubes are pin-for-pin drop in mechanical and electrical replacements for the 6J1. Sometimes a tube might be listed as a substitute, but could have different electrical characteristics or pinout connections.
If the 5654 work out , here are some joint army navy tubes that should fit the bill .... Bill
https://www.amazon.com/Matched-Upgrade-AMPTATA-Fully-Tested-Platinum/dp/B0BM4HJ5YV/ref=sr_1_1_sspa?keywords=ge+5654+tubes&sr=8-1-spons&sp_csd=d2lkZ2V0TmFtZT1zcF9hdGY&psc=1
I've only opened and tested one kit so far. No missing parts. I tested all the resistors and they all appear to be very tight tolerance, 1 to 2% of nominal value. All 8 of the 470uF caps measure about 445uF, about 5% low. They sent four 1uF ceramic disc 50V coupling caps. They test OK, but I am going to replace them with some small 64V polyester caps that I have on hand. I think you replaced your 1uF coupling caps as well. All transistors and diodes check OK. The 50K pot measures 50.8K and seems to track well channel to channel. So far, so good.
Tom, I see you substituted Nichicon ES series 3.3uf bi-polar caps for the input and output coupling caps. Looks like a really tight fit next to the resistors. I found some WIMA 3.3uF 50V polyester caps in my unused parts box. They are about the same size as the Nichicon ES 3.3uF's, but they have very short leads. So I will have to solder extensions onto the existing leads so that I can mount them slightly proud of the resistors underneath them.
https://www.mouser.com/ProductDetail/WIMA/MKS2B043301H00KSSD?qs=ISKKXg98%2B1m7nRRaqaljlg==
Bill, the 5654 is a ruggedized version of the 6AK5, from what I have read. I have them running in mine now. I'll have to double check, but I think the distortion is a little higher with the 5654 tubes. But I don't hear any sound when I tap them like I did with the stock tubes.
Glad you weren't missing any parts! I know there had been some complaints about missing parts on some of the Amazon listings, but there seem to be a few different sellers of the same basic kits. I was going to put some Wima or Kemet film caps in place of those ceramic couplers, but like you said, it's pretty tight between the resistors. I have a ton of the ES caps, so I thought I would use those instead. They sound and measure fine to me.
I was curious, so I hooked up the heaters directly to the little 12.3vac transformer. There is some hum, but it's pretty low when listening with headphones. I'm just guessing, but I think the AC is being picked up on adjacent traces somewhere. I'll still end up using a diode, cap, and the little regulator board.
In college I built a guitar sustain unit as a final project. I put a ceramic cap in the 1st stage and it worked fine on the bench but if you tried to use in the real world it would scream with feedback. Ceramic disks can be very microphonic and I would never use one in an audio circuit.
Ron
+1 to ceramic caps sucking. Only thing worse in my experience is tantalum caps.
I tested my set of four 6J1 tubes on my Heathkit TT-1 tube tester, which can measure transconductance (GM) in micromhos. I used the 6AK5 roll chart data to set up the tester, so this may not be entirely accurate for a 6J1 tube.
Kit 1, Tube A: 2800 micromhos
Kit 1, Tube B: 3550 micromhos
Kit 2, Tube A: 4500 micromhos
Kit 2, Tube B: 3450 micromhos
So, it looks like I lucked out by ordering two kits. I can get a good match by using the 3550/3450 pair.
EDIT: I just measured the heater voltages in the final series circuit and the way the voltage splits between the two tubes appears to be closely related to the measured GM value. When I plug the matched set of 3550/3450 tubes into the sockets, I get the same heater voltage at pins 3 & 4 of each tube: 6.15V. But when I plug the unmatched set of 2800/4500 tubes into the sockets, I get a heater voltage of 5.8 volts for the 2800 tube and 6.25 volts for the 4500 tube. This makes perfect sense, because transconductance goes up when the filament burns hotter.
Wow - the pairs packaged with each kit are pretty far apart. Glad yo got a decent pair to work with. I'm not sure how much variance is enough to cause an imbalance in this design, but now I'll be testing mine too.
I finished populating my main signal path PCB. I added some jumpers to convert the board from a dual polarity +28, 0, -28 volt supply to a single polarity +58, 0 volt supply. Instead of using my second PCB for the other power supply parts, I decided to make a small custom circuit board from scratch. This was much easier than modifing my second PCB because that would have required cutting several circuit board traces and I didn't want to do that. So now all I have to do is install the two boards in my main cabinet and power it up. I put the rectifiers and first RC smoothing stages next to the transformers.
Looking good! What did you use to etch your PC board?
I used a small PCB prototyping board to lay out the hole spacing and then taped this prototyping board to a single sided board. Next, I drilled all the holes with a #60 drill bit using my drill press. I then buffed the copper to a high sheen with #0000 steel wool and cleaned the surface with rubbing alcohol. Next, I painted all the traces free hand using a king size Sharpie. After gloving up, I filled a small plastic tray with a small amount of ferric chloride and set the board in the tray, copper side up. Using a small, wooden stick, I lightly agitated the board back and forth until all the copper was removed between the traces. This took about 20 minutes. To stop the etching process, I rinsed the board in running water.
I then cleaned the Sharpie ink from the traces with rubbing alcohol and disposable paper towels. To keep the copper from oxidizing, I tinned the traces with rosen flux and a small amount of solder. You can also tin the traces with a tinning solution, but I didn't have any tinning solution on hand.
Very nice! I've never etched my own boards, but always wanted to try it.
Aw yeah. That's gotta be some ASMR Sh@t tinning those traces!
I tried that twice back in the day when Radio Shack sold kits. Never worked, so I never tried it again. Yours look great!!👍🏼
I built a few pcb's back when Radio Shack had those kits. Also had a construction class in college and we had to design and make our own pcbs. Funny, I really loved that class
I was once set up to screen print PC boards and etch them. I got tired of drilling all the holes so I would send my artwork off to an outfit that would make the boards for me. I had a cad program that was dedicated to PCB layouts and once did a custom layout for a audio outfit here in Iowa. I still have equipment with some of the boards I made.
I used to roll my own preamps, power supplies and active crossovers.
I have all the parts for the chassis, I think. But I need to put my project on hold for a bit. I have a vintage receiver restoration for a paying customer on the bench.
These kits came with a pair of small RCA jacks that mounted directly to the back of the PCB:
But since I decided to mount the boards inside in a huge cabinet, I decided not to use them.
Instead, I removed the spring loaded speaker jacks from the left corner of the cabinet:
And replaced them with with RCA jacks mounted on a custom board:
Later today I'll mount the transformers, on/off switch, 3 boards, and wire it all up. Should have it playing music sometime tomorrow.
Woohoo! Success! Powered it up with dim bulb and variac late last night. No smoke or fuses blown, so I gave it full power. My targeted B+ of 56VDC actually came out to 58.3VDC (close enough). My targeted heater voltage of 6.3VDC actually came out to about 6.15VDC on each tube (close enough). Ripple on the B+ and heater supply was unmeasureable on my scope, probably somewhere in the 2 to 4mV rms range. There was no hum at all from my bench speakers, even with the volume turned all the way up and my ear next to the woofer cones. I connected the outputs to my small Lepai 2020, which has an input impedance of 10.9K ohms. Driving into this load, preamp gain at 1kHz was 11.19dB in one channel and 10.88dB in the other channel. I played several music selections from my old cell phone and it really sounded good. So far, so good. Now I can put the front panel and top cover in place and do some frequency response measurements. This preamp has a very high output impedance, so I want to test to see how badly the high frequencies roll off when driving into a 3 to 6 foot long RCA cable and connected to a 10k ohm load.
Looking good, Bill! Your power supply is probably the biggest ever used with this little kit. I'm curious to see your measurements. Have you done the feedback and other mods yet? I see the Wima caps.