• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Van der Veen 70W tube Amp DIY

Jump to Latest
#21
2e upgrade

I also did 2 other mods ....

1) A 2uF/900V Polypropylene (from Rarefaction in Melborne Australia) directly across the high voltage at the Output Transformer Centre Tap connection to 0V.
2) A 100K 3W Metal Film bleed resistor across that cap.


the high voltage of the output trannie andto the centere tap to 0.V are which colour of wire.....

Think your upgrades are very useful for the stability and sound of the unit but i take a 2 or 4 mfd paperinoil instead.
 
#24
kathodyne said:
Hello all....


www.mennovanderveen.nl (english)

go to 'New' -> 'modify the UL40-S2 step by step'

gr.
Click to expand...


It looks a bit of a GLITCH from the Menno site ...QUOTE ""Did I make a mistake in my transformer design? No, I didn't. These good low frequency results only show that I am continuously improving the VDV-GIT80 transformer. Especially a new high quality core steel is responsible for this fine result."" um...

Steel in transformers looks new to me...:drink:I'll love to see the BH curve to prove it isn't mouldy old silicon iron with some cobalt thrown in. Any of his ideas are new ? Nup.

To get 100W output from a pair of EL34's would put'em in the slaughter house, but if he does it then I suppose there must be a darned good reason for doing it. Not for me!

richj
 
#25
To get 100W output from a pair of EL34's would put'em in the slaughter house, but if he does it then I suppose there must be a darned good reason for doing it. Not for me!
Click to expand...

Me neither! According to some manufacturers' data, you can get 100w from a pair of EL34s in class B pentode mode, with very high plate voltage of 800v and screens at 400v, but that's only for PA type use.
 
#26
Re: Yep - I built a pair just recently

gingertube said:
I built the 100W versions using 4 x Winged C Svetlana EL34s.

I started off with them in Ultralinear Mode and found I had two problems.

Problem #1 - Slight bursts of HF Oscillation just after the postive peak of a sine wave. This was cured by increasing the 150R screen resitors to 1K. I reported this to Menno and he has put a note on his web pages to the effect that this mod should be "Standard".

Problem #2 - The Output Impedance of the basic amp (no global feedback) is 12 Ohms. There was some "colouration" of the sound as the output power followed the speakers impedance curve.

To cure problem #2 I rewired the output into triode mode. Output Impedance dropped to just less than 5 Ohms. Output power dropped to 40W BUT full power bandwidth is 12 Hz to 150 kHz

If you build the 70W version you can bias the output tubes a little higher and this will drop the output impedance to probably about 4 Ohms or just less.

In this arragement these mono-blocks have become my favourite amp(s). I have an 845 SET and a Hugh Dean AKSA 55 Nivarna Plus (Solid State) which these amps have displaced.

Have a look on the Photo Gallery above about the second to last page for the Gingertube "Menno Bricks" for some pics.

P.S. My speakers are VAF Research DCX s with a nominal impedance of 6 Ohms (95dB sensitivity). If your Speakers are less than 4 Ohms or if you are using Electrostatics then the Output impedance will probably be a bit high.

Oh - one last thing I pulled the 12AU7 and am using ECC99 in its place. I thought that gave me a slight improvement.

One problem with the design is that the Output Tube grid resistors are reasonably high and is in fact too high for anything except EL34s. Don't be tempted to use 6550s or KT88s. In any case the tube separation really only suits the skinny EL34s. I haven't checked data sheets to see if the KT66 that someone above suggested is a SAFE substitution or NOT.

I RECOMMEND that you use the "Winged C" Svetlana EL34s and not try to substitute cheaper alternatives.

I also did 2 other mods which I have no idea if they helped or NOT.
1) A 2uF/900V Polypropylene (from Rarefaction in Melborne Australia) directly across the high voltage at the Output Transformer Centre Tap connection to 0V.
2) A 100K 3W Metal Film bleed resistor across that cap.

I suggest that you put the electrolytic caps on the back (the tracks side) of the PCB just to keep them a bit further away from the heat of the tubes. Also put the bias pots and the bias measure resistors in each output tube cathode on the back of the board so that when the board is screwed down into its chassis you can get at the bias measuring points and the adjustments.

Due to the 250V AC Oz Mains Voltage my rail voltage was 480V instead of 450V so I backed off the OPutput Tube bias currents from 50mA to 45mA to keep anode dissipation at 22.5 Watts.

Good Luck
Ian
Click to expand...

Ian... To cure problem #2 you rewired the output in triode mode ....can you tell me how it can be done ....which colours to take
I have changed the grid resistors to 1K allready but need your help with problem #2
 
#28
excuse me for the non-intended thread jacking......

i saw ul40-s2 mentioned over here....and thought to be helpful...
the comments about the transformer core mentioned by richwalters.....have nothing to do with el40-s2, nor the vdv 100, the vdv-git80 is used as versatile transformer in menno's 'The Project' (a matrix of different topologies, of which 20 have been build, look closely.....)
The dutch word 'kernblik' does not translate into english(would give core-can), reason why the word steel is used, i guess......

@max29: you're right....i should have posted another thread....

about the vdv100, i've listened to it, at menno's, who built it around 1987.....he said he wouldn't do it like this anymore....he could even redesign it, but he doesn't because it is an example of history, of a learning curve.....

I didn't really like the sound of it, although it really had the 'oompfhh' factor (you notice very well in the sound that there are many watts 'ready to go' which is nice with dynamic passages in music)

gr. kathodyne
 
#30
I've uncovered an old quote to British Standard BS3860 (pre 1967) that any amplifier built to this conformity is unconditionally stable under EVERY operating condition including o/c, short circuit capacitive ouputs and long speaker cables. And more to th point, that includes UL operation.
So if my forefathers got the designs to run right, by now we shouldn't be having so many problems.
It seems tube amp design problems using Plitron transformers never ends. When compared to EI laminated cores which have set standards why not settle and be content ?
To me, hinting at smps push-pull currrent mode forward technology, a Plitron core run on a cycle per cycle current sense basis is the only viable design solution when the full range BH characteristic to zero is used. It seems only logical that when such a core is faced with transient conditions, it will immediately saturate. Anyone tried designing around it ?
No names given, but I've gotten right off tube amp designs which haven't faced the laws of physics.

richj
 
#31
Well, I have never used any of the Dutch gentleman's transformers and cannot criticise (neither do I); measurements certainly look impressive. But I also seem to have noticed enough problems in correspondence over the years to make me wary: Were these definitely transformer-related, or was the circuit suspect?

I myself use C-cores and can certainly not complain: Their use in a 120W tube amp (pri. L=125 Hy, power bandwidth going down to 100W in UL at 15 Hz, lower level down to 2 Hz) to my mind gives little reason to quibble. The steep slope of the "threshold-of-audibility" graph below 30 Hz makes 100+W there mandatory in any large living room before you can hear anything (plus the loudspeakers that can handle it). These days with strange noises below 20 Hz easily accomplishable, perhaps make me old-fashioned; but I am happy with say 30 Hz. In that sense and with reliable predictability, I personally also would need some very good reasons (and assurances) to change to toriods.

Any other makes use them for output transformers?
 
#32
Conversion to Triode Mode

Max29 (and others),
Sorry about the tardy reply - I've been away for work, just back.

I'm using the VDV2100 (PAT4006) but same goes for PAT4004.

The PCB shows connections for the primary of the Output Tranny:
Black (Anode Push Side)
Violet (Screen Push Side)
Red (HT+ Supply)
Brown (Screen Pull Side)
Green (Anode Pull Side)

To convert to Triode Mode
Disconnect Violet and Brown wires - heatshrink (double layer) over the ends of these wires and tie them back out of the way.
Link the PCB point where the Violet Wire was to the Black wire connection.
Link the PCB point where the Brown wire was to the Green wire connection.
Thats all you have to do - Output Impedance is still a bit high (approx 5 Ohms) but sound is very good with outstanding detail.

My "Menno Bricks" are sitting on the shelf until I get around to rebuilding the front ends. The current plan is to rebuild these with the front end shown in the 6V6 version of the "Baby Huey" design which is posted here.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=72536&perpage=10&pagenumber=6 That is I am going to apply a bit of shunt feedback to get that output impedance down to around 2 Ohms.

Don't obsess over the fact that I intend to redesign / rebuild the front end on mine. The existing simple Common Cathode plus direct coupled Concertina is fairly good, especially if you sub an ECC99 for the 12AU7. The Concertina splitter actually has extremely low output impedance and drives the output tube grids well (until you hit the Class A to Class B transistion of the output tubes - that is while the loads on the anode and cathode remain equal).
Cheers,
Ian
 
#34
Triode Mode with Amplimo XC462

max29,
To convert to triode mode when you have the Amplimo XC462 Output Tranny.

Disconnect Pink wire, heatshrink over end(two layers) and stow the wire out of the way. Link the PCB connection from where the pink wire was to the white wire (anode lead).

Disconnect Grey wire, heatshrink over end ( 2 layers) and stowe the wire out of the way. Link the PCB connection from where the grey wire was to the blue wire (other anode lead).

Re check bias of output tubes after the conversion, it will be a bit different in triode mode than it was for ultralinear mode.

Cheers,
Ian
 
#35
Update,
I haven't used my VDV70100 "Menno Bricks" for about 12 months as I've been concentrating on The "Baby Huey" design and a few other projects.

Dragged them down from the shelf last night - checked the bias and hooked them up. They really are quite seriously gorgeous.

Pulled the Winged C Svet EL34s and put in a set of JJs, still sounded good but the Winged C Svets are better (in triode mode at least). Back to the Svets - I'm going to listen to these monoblocks for a couple of weeks before going back to other projects. The comparison has already showed me that I have pushed the feedback in my Baby Huey too far and that the original values I posted on that thread are better. I was trying to get too much damping and sacrificed pace, impact and slam to get it.

Just goes to show you that its always a good idea to pause your development work every now and again and cycle back through all your amps to check where you are at - and possibly more important what you are aiming at..

Aside, lighthearted advise:
Always aim high - you might still miss your target BUT at least you won't shoot yourself in the foot.

Cheers,
Ian
 
#36
Can i replace an 1N4007 diode or the byv 26E with f.e. an hfa 1200v 8 a diode.

I do not like the sound/signal from an 1n4007 a.o. 1N-diode at all and is is rather harsh and cold and even the better byv26E is non-audio minded stuff...to my ears.

So can i replace them with an fairchild ultrafast and softrecovery diode like the hfa08tb120
 
#37
With all of the talk of Van der Veen's toroidal P-P OPT designs, I was wondering if the same reservations some members have about them extend to his SE OPTs?

Given that they have such outstanding specs, I'm surprised that they never seem to come up in the usual Tango-Tamura-Magnequest discussions. The prices from Amplimo are fair in this kind of company. I am under the impression that the SE toroid OPTs could outperform some of these... Any experiences?
 
#38
Max29,
Yes you can certainly replace the rectifier diodes with any suitably rated ultrafast soft recovery power diodes. I had a look to see if I'd done that (as I usually do that to all my amps routinely) and found that I had not BUT had fitted 22nF/2kV ceramic caps across each diode which would help with rectifier switching noise.

I actually recommend that you do both - that is fit ultrafast soft recovery diodes and put a 10 to 22nF 2kV ceramic capacitor across each diode.

There is no excuse for using 1N4007 in any audio project these days - A visit to my local electronics store recently scored me a reel of 200 ultrafast soft recovery diodes with ratings the same as 1N4007 for AUS$10 (5 cents each).

Cheers,
Ian
 
You must log in or register to reply here.
Top Bottom