• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Tubelab Universal Driver Board, 2015 version

Which Rigol oscilloscope did you purchase? For working with tube amps will 50mhz suffice? Are two channels enough? Are there any other parameters I should be concerned with?
thanks, Jacques

I bought the 50MHz DS1052E. The display is a bit small, the noise floor could be a bit lower and I had to buy another 2 probes because one of the ones that came with it was a bit dodgy. As noted I also bought a high voltage probe. The probes aren't expensive to I should probably buy a second.

All that said, it's amazing what you get for your money ($US259 at the moment I see). It will do FFTs, has all sorts of maths functions like adding and subtracting traces and can display all sorts of waveform information on screen. With the USB connection you can control it from your computer and have the traces displayed and saved there.

OK, I'd love a big, expensive professional oscilloscope but this one meets my needs.
 
I got a Hantek DSO5072P at the end of Spring. 70 MHz, two channels, decent display, decent probes, seems easy enough to use. It does math and FFT, but at only 8 bits, so only good for high level signals - 50dB or so. eBay had a flash sale and it was $203-204 delivered to my office. It's a lot of instrument for that price. You will want at least two channels, so you can look at the input and output of something simultaneously.

I like the new Siglents a lot, they have spectrum shading displays on even some of the low end models ( I think some Rigol's do too ), so more like an old boatanchor CRT scope. If I were to do it again, I might cough up a few more bucks for one of those. I try not to spend much money on hobbies, though. The Rigol's look nice; I am continuously talking myself out of one of their inexpensive spectrum analyzers. The low end Tektronix DSO scopes looked like too much money for not enough scope for hobby use. Ditto for some of the other professional lines.

I notice a lot of the DSO's now have a function / signal generator in the box. That could be useful for signal tracing to have that built in - saves the cost / space of another instrument, but likely too high distortion/noisy for much audio use.

I'm not that certain a wide bandwidth is all that necessary for most AF hobby stuff. I got a handheld 1 MHz ProTek DSO about a decade ago when it was new technology, and it immediately moved all my boat anchor stuff to my warehouse, never to return. I have one of the new DMM's with a graphic display, only 15KHz of bandwidth or so, but I still find that useful. It has moved most of my DMM's off the bench.

If you are fooling with tubes, you want to make sure the scope can handle the voltages. My ProTek is good for 600 volts on the inputs. Not sure about the Hantek - it has only seen low voltage RF so far, that is the only reason I bought it.

I am just a hobbyist, so my opinion / needs are strictly from that perspective. Professionals will have a much different opinion, I am sure. If I could make a business case for one, I would get a TEK and depreciate it. But I can't.

Win W5JAG
 
The probes determine the voltage the scope can handle.

Unless they don't use probes. Probes are a hassle for a scope that may go to the airport, car shop, field day, etc.

If the old Heath SSB monitor scope can be counted as a scope, in my current inventory, probeless scopes outnumber probed scopes 3:1. I think I have an unbuilt 5 MHz Heathkit in the attic, so that would make it 3:2 if that ever gets located and built ( not likely ) .....

Win W5JAG
 
That is true, however most people on this site will not encounter a probeless scope.

Therefore it is good advise to recommend they purchase HV probes and avoid destroying a new toy when they hook a 350V probe to the plate with 250V B+ applied and damage the input with the near 500V Peak waveform at full output.
 
Well I'm disappointed, Finally cranked it up to 350V B+ with a pair of 6P3S tubes (after testing 47 of them) and I get 5Vp-p out clipping on the 8 ohm tap to a 8 ohm resistor with 400V plate to plate and red plate one of the 6P3S tubes.

Tried a second CXPP45-MS-6.6K and got the same results.

Swapped to an XPP15-8CT-7.6K and got roughly 17W at 1KHz with no problem.
 
Last night I tried a CXPP30-MS-5K and got 27W with B+ at 350V. Noticeable distortion of the waveform occurs when transitioning out of class A into class B.

I am not running any feedback at this time, and full output before clipping requires about 0.35Vrms.

Tubes are 6N1P-EV and 6FQ7/6CG7.
 
Last night I tried a CXPP30-MS-5K and got 27W with B+ at 350V. Noticeable distortion of the waveform occurs when transitioning out of class A into class B.

I am not running any feedback at this time, and full output before clipping requires about 0.35Vrms.

It'll be a while before I have hardware to test but according to my simulation (6CG7s and 6550s, 5K Hammond OPT, 50mA bias current) your results for a 350V HT are in the ball park.

The first image is THD vs Power for 350V. The second for 500V.

The other 2 curves are for the anodes of the first and second LTPs.
 

Attachments

  • 350V HT.png
    350V HT.png
    60.9 KB · Views: 342
  • 500V HT.png
    500V HT.png
    66.1 KB · Views: 326
Whoop whoop, definitely throwing my hat in the ring for this one. Super interested in building an amp with a pair of the UDB boards. I think I might finish rebuilding my "Industrial SSE" then skip the SE board and go straight to the UDB. Already impulse purchased some 6CG7's...

Either way, restarting from first post in the thread and shall ask questions when I make it back around to the end. Just deciding whether or not to let my itchy trigger finger buy the UDB boards prior to the PSU information and PSU boards themselves becoming a little more clear...
 
I had to pull the board off the workbench for other projects for now. Spent the day pressure washing the driveway cracks and filling them with crack filler.

It will probably be a week or two before I get back to playing. Summer is winding down and the garden needs work along with trying to get some work done on the house before winter.

Two years till I sell and move on.
 
I decided to go ahead and invest in a Rigol DS1045Z. My plan is to use it to the scope to maximize the settings on the UDB board.

I am using 6CG7’s and KT88’s. Per George’s startup instructions I have set the plates in V1 at 110 volts and the plates for V2 at 450 volts before installing the power tubes. I am used a signal generator ap on my iPhone set at 1000Hz with a dummy load.

I am assuming the goal is to adjust the bias settings and the V2 plates to keep the sine waves from distorting.

Questions:

1. Do I need to purchase a real signal generator or can I get by with my iphone ap?
2. What is the best method for making adjustments using a scope?


Although I’ve looked at a couple of articles and YouTube videos about how to use the scope I’m learning by doing. Any suggestions would be most appreciated.

On the bench with these settings the amp sounds great

Thanks, Jacques
 

Attachments

  • IMG_2646.jpg
    IMG_2646.jpg
    788.8 KB · Views: 494
  • IMG_2647.jpg
    IMG_2647.jpg
    961.2 KB · Views: 491
Stress to R6 & R7

I've had two udb monoblocks (Hammond 1650R, KT-88) up and running satisfactorily since mid July. During that time the driver boards have been mounted on a piece of plywood, socket side up. In disassembling the first amp to transfer it to the final chassis, I removed the udb and inspected the under side. Clearly the R6 & R7 (2W, 43K) resistors are damaged. (A new resistor is shown alongside.) Also, the solder joints seem iffy as well. Of course I have used these two resistors as test points for setting voltages throughout building and testing; so perhaps that testing has been stressful.

Anyway, I need to do something about this - at minimum replace them with heavier resistors (3W, etc.) and refresh the solder joints.

Comments appreciated.
 

Attachments

  • r6-7.jpg
    r6-7.jpg
    685.8 KB · Views: 466
at minimum replace them with heavier resistors (3W, etc.)

The resistors in my board don't look toasty at all, but there are lots of variables at play here and yours are obviously not happy. If you have about 450 volts of B+ and maybe 110 volts on the plates of V1, then there are 340 volts or so across the resistors. The usual bias current for V1 is about 10 ma split between the two halves, or 5 mA per resistor. This makes each resistor dissipate 1.7 watts, which is a bit close for a 2 watt resistor, but it works in my board. Not all boards run on 450 volts, or 10 mA through V1.

My boards have the resistors on the top side, and they are spaced about 3/8 inch above the board, so the heat can get out. Your resistors are mounted pretty close to the board and have limited airflow. Heat rises, but it's blocked by the board, so it has nowhere to go but build up in the resistor, the solder joint, and the PCB.

I would got to at least a 3 watt resistor, and space it away from the board. Mouser has some 5 watt metal oxide film resistors, but 43K is not stocked. They have 39K and 47 K in stock.

The current trend is toward resistors that are much smaller than an resistor of the same wattage rating that was made just a few years ago. The laws of physics haven't been repealed, stuff 2 watts through a resistor that's half the size of an older 2 watt resistor, and it will run hotter.

Modern high temperature parts can take the heat, and modern no - lead solder takes more heat to melt, so all looks good on paper. Trouble is that none of the resistor makers even dream of building tube amps where several resistors and some hot tubes are all stuffed into a box. They stick one or two into a power supply......

As you have seen the hotter resistors will transfer heat up their leads to the solder joint......it's even part of the new resistor's temp spec. The solder joint and PCB is part of the resistor's heat sink.

For this reason I allow as much lead length as possible to transfer less heat to the solder joint. Heat rises, so if you are putting the resistor on the bottom, space it a half inch or so below the board. If given the choice of two different 3 watt resistors, go for the larger of the two. The extra surface area will help shed some of the heat.
 
Thanks for the explanation. FWIW, the resistors are mounted higher than they appear in the photo, but being on the underside with a solid surface an inch or so away they're getting a lot of trapped heat. I ordered some 3W, 43K replacements from Digikey; and on the final chassis the resistors will be open to the air. Hopefully those two adjustments will do the trick.

For info, along with a couple of friends, we displayed the amps with a diy pre and speakers at Burning Amp in SF. The system was well reveived (although the Nelson Pass exhibit managed to get most of the attention.). :rolleyes:
 
George...Excellent excellent excellent!!!


I've been distracted by a camper van build and a camper van forum for the last 5+ years but I've been thinking about building a triode PP KT88 amp with some really nice iron I have stashed in the closet.


My thoughts have been to build sgregory's mosfet follower Mullard design, but Chrish's amp with the LTP's keeps calling to me.....and at this point you've got some driver boards for just that......I'll take a pair....


George, I just sent you a PM about a paying "work" project as well......
 
Last edited:
I finally made it through the 6L6 AB2 thread and this one.

One of the interesting things I picked up assembling the uTracer a couple of years ago is placing ceramic beads on the leads of a large resistor to raise them off the board. HP was fond of using them on ceramic capicitors to keep them from flopping over.
 
I finally made it through the 6L6 AB2 thread and this one.

One of the interesting things I picked up assembling the uTracer a couple of years ago is placing ceramic beads on the leads of a large resistor to raise them off the board. HP was fond of using them on ceramic capicitors to keep them from flopping over.

I picked up on that too. After searching for reasonably priced ceramic beads recently I ended up buying 2700pcs/lot White Glass Seed Beads Bulk 2.5mm Spacer Beads off AliExpress for US$2.75.

They're not exactly white but they'll do the job.