To view high resolution images, right click on image and select View image or Save image
Huldra 10 where first seen in the catalog for 1972-1973 together with TR-1000 and TA-300. How Huldra 10s amplification compares to TA-300, TR-200, TR-1000/1010 and later TR-1040/1055 can be seen here.
This Huldra 10 where received as partial functioning. General cleaning of all switches with Arrow Lotoxane and oiling of the axel for AM tuner where done.Axel where really stuck and afer some jiggeling what made it come loose where to remove the E-ring on the axel and then press the axel down into the houseing of the variable capacitor while turning back and forth. Take care not to damage the tiny fragile brass and aluminium gears. Had to shorten the AM dial cord 2-3mm to get better grip.
Mains wire for the transformer where moved from the 220V tab to the 240V position (outer tabs) and 25V PSU adjusted accordingly (VR R1002). Actually after the primary winding change voltage dropped to 25.1V.Quiescent current where now low around 5-6mV after 15 minutes warmup. Adjusted to 11mV (60 mA) according to service manual. There are some that recommend turning it a notch higher to 18mV (100mA) for a cleaner sound. Better familiarize myself with stock setting before setting out on that journey.
"One of the best World-Band Hi-Fi Receivers ever!"
Original 4700µF capacitor where replaced with a 10.000µF based on multiple recommendations. This give measurable lower ripple noise and bass reproduction benefit from it. The fuse and rectifier are capable to handle the added inrush current at power-on.
A new hole had to be drilled for the large cap. Actually, the 10.000µF cap are smaller than the original, but have a different foot print. To ensure correct spacing I used the new cap as a compass inserting one pin to my designated hole and scraped an arc on the area where the new hole where to be drilled.1.5µF/100V film cap where soldered on underside of PCB over the big cap terminals using free holes left from the original cap. Small and large caps combined supposedly give a better broadband circuit having different self-resonace frequencies.
Only a single bulb out of eight 6.3V/320mA E10 bulbs for the radio scale panel where lighting. Rather than replacing all of them with new bulbs the idea to use LED open up the possibility to have this powerful 5.8V secondary transformer winding powering a Raspberry Pi (RPi) mini computer to give Internett radio in place of the now closed Norwegian FM radio network.
White LED where first tested with decent result. Front glass on these old Tandberg receivers are tinted blue. While they normally fade when exposed to UV light this specimen still have a clear deep blue color which really pop when scale where backlit with blue LEDs. For AM scale there are now total of 16 LED and 6 for the FM scale. Huldra, the old dame emblem will have her own LED up her back — planned to be lid up when there are an Internett connection.
While LED functioned fine on AC current flickering where notisble. An 3 ampere rectifier where installed clamped to the top edge of the chassis, right over the main rectifier. With a 1000µF cap the new circuit now output 7.8V DC with no load.
Note: Brown wire to center tap for 4.2VAC indicator bulbs are removed from transformer side, taking these bulbs out of circuit. Later they be feed with positive DC current with the Brown wire connected to negative terminal on this new PSU.75Ω resistor where used for each pair of LED. Tests after installation now show we can add at least another 75Ω in series with the finished LED strip and still have the blue come trought even in bright daylight. While bright blue looks really cool it soon gets tiresome — dimmed down is proper way to go one think.
A LM350 voltage regulator where installed after the new rectifier (for the LED) before I understood more tests should have been performed beforehand. As it turn out voltage drop are considerable from idling at 7.8V DC. Test now unveil that without voltage regulation I can only get some 700mA/5VDC. This would need a LDO (low dropout regulator) rather than the LM350 regulator to be useful.
Loading the rectifier harder with a 2.2Ω resistor voltage now drops down to 4.09VDC ≈ 1.9A.
To keep these figures indicator bulbs also have to be converted to LED, or removed from circuit with it's total load of 575mA when all bulbs are on.Not sure this will perform as planned with RPi B3+ needing close to 980mA at full throttle, — idling at some 350mA. Then add a dedicated DAC+ DSP hat from HiFiBerry system will work close to the edge with regards to keeping voltage within specification.
emitter resistor value * quiescent current = voltage over emitter resistor
For Huldra 10 the default setting are: 0.18Ω * 60mA = 10.8mV ≈ 11mVmeasured voltage over emitter resistor / 0.18 = quiescent current
"Closest I came to tube sound without the hassle of tubes"
"Tandberg did well with the transition from tubes to transistors"
anonymous forum members
Not planning to use this Huldra 10 with any wintage tape machines bringing Line Output levels up closer to modern specification seems a good idea. Current configuration has 40kΩ output impedance and 5 times lower output gain than most modern gear. Higher output will prove very useful if Huldra 10s RIAA/phono amplifier are to be used as a stand alone device.
AUX Out — Remove R561, R562 (47kΩ, leave emty) and replace the two 330kΩ resistors R567, R568 with 10kΩ.
TAPE Out — Remove R565 and R566 (47kΩ, leave emty) and replace the two 330kΩ resistors R571, R572 with 10kΩ
PCB revision 41160-6 found in this reciver the resistors R567, R568, R571, R572 where 280kΩ
For hi-fi gear there are some that recommend vintage carbon resistors in this positions. For period correct resistors 10kΩ values can be fount in the AM tuner at R319, R329, R335, R336 and a few others. They could be replaced with modern resistors, or just removed if AM tuner aren't in use. I chose to replace them. Not finding R319 instead snagged R313, part of a divider network I got a little jittery ending up with a burned transistor if simply removed.
One aberration for the line out mod could be to mount RCA sockets with it's own 10kΩ resistor and leave the 330kΩ in place. Then mount a switch to the 47kΩ resistor going to ground (those we removed). This setup will then let us hook up and configure the DIN socket for vintage gear switching in the 47kΩ, and use RCA for modern without the 47kΩ to ground. With only one device physically connected at any time thought. Schematics for TR-1000 actually show similar setup, but the 330kΩ resistor are 150kΩ with 47kkΩ always in circuit to ground with seperate 10kΩ resistors inline for the RCA sockets.
This concept that have been a known way for studios to record electric guitar and bass amps since early years of recordings where first advocated used with Tandberg receivers by Kjell B. Whom also recommended lowering the quiescent current to 6mA (1,1 mV over R 721-722) when used as headphone/pre-amp with no speakers connected. This should have the output stage work in Class A up to 200 mV from where it become Class AB. Although there are reports that crossover distortion might become a problem when lowering the quiescent current this low.
This last setup has actually been reviewed on print back in 5/2004 Fidelity #10. The article where an exercise comparing vintage 70's Huldra 10 to at the time modern Krell KAV-400xi to learn if the world of hi-fi had moved forward. H10 as an amp with original power supply fell apart due to low wattage for the test setup, but as an preamp it held up remarcable well.
TIP: This technique can possible be used as poor man's Neve/SSL style color box in your recording studio. If you favorite amp are missing headphone out use a load box for guitar amp. Or make one as shown in the drawing above. Higher wattage amp would of course need resistors capable to handle the power.
The initial idea where to find a way to shut down any oscillators when H10 where used for amplification only. Several hours were wasted to trace main PCB. Sadly revision 41160-6 PCB layout does not show up in any of the service manuals I found online.
Interesting I found the screening bar going over the PCB front to back do not all go to ground as expected. The one separating PSU and preamp actually feed 25V to one part of the AM tuner. But the actual AM oscillator takes 25V from the push button switching system in front. So does the FM tuner.
As it turns out Tandberg had the same idea that turning off the radios completely is preferred. This can be verified by watching the small bulb inside the 25V power supply. The more current the system draw the more this bulb will light up. From barely glowing when any of the line level Phono/Tape/Aux are selected. To very visible when any of the AM band are active. And full glow when FM with presets are selected.
|Fill inn the green box to experiment with various capacitor values. For instance are 2200µF used in some Sølvsuper receivers.|
C719/C720 coupling capasitors -------------------------------- Although not being particular accurate applying calculations fc=1/2πRC for a RC high pass filter using a typical speaker resistance as the resistor value tell us there are negligibly difference in cutoff frequency between the two capacitor values typically used on the power amp outputs. 3300µF 4700µF | 1000uF 4Ω 12 Hz 8.5 Hz | 39.8 Hz 6Ω 8 Hz 5.6 Hz | 26.5 Hz 8Ω 6 Hz 4.2 Hz | 19.9 Hz Understanding this less than adequate and simplified calculation we antici- pate that the real speaker impedance will influence about equal for both cap values.
But what differ the most are how these switches are implemented electronically. Where each switch either select or pass on the signal. Meaning if one input is not working properly the contact problem could possibly be located on a switch further up the signal path not passing on the signal as intended. For instance the Phono switch select between Radio or Phone. Which are then passed on the Aux switch and from there to the Tape switch before going to the volume control and preamp.These switches are actually designed to be taken apart and serviced. Although air pressured cleaning fluid are the popular way to do these repair. My absolute favorite are Lotoxane with its long open time let me soak and dissolve old grease and oxidation without drying out and create a new layer of dirt on parts and contact elements. I use the liquid version and apply air pressure afterwards from separate container.
Capacitors order list for Tandberg Huldra 10 -------------------------------------------- All capacitor are electrolytic type unless noted. I like to keep as many original components as possible. But for safety reason and possibly being those caps that had most stress through nearly 50 years I replace all electrolytic caps on power supply and power amp except the input coupling caps which are some special selected types compared to power rail caps with identical value. I'm unsure all 2.2µF used to block DC on inputs on each module truly are tantal type. Dielectric are unknown and it might be better to call them drop capacitor. They do have similar mushroom shape and dipping process into liquid insolation layer as typical tantal does. Anyway, they are part of the sound and we better keep the originals if possible. If replaced try film capacitors. NOTE: Several 22µF values have 47µF mounted on PCB on newer revision without schematic being changed. Power Supply #41150 C1006 - 4700µF 70V (*) C1005 - 220µF 50V C1004 - 10uF 100V (35V) C1003 - 22uF 40V (47µF original on PCB) C1001 - 100µF 50V (feed stereo indicator bulb)(**) * rectifier and fuse seem to hold if increased to 10.000µF which give better ripple measurement and bass performance ** Tantal on PCB but not on schematic parts list Power Amp #41115 takes 61V from PSU C725 - 470µF 63V possible has to do with power mute circuit Left: C701 - 22µF 63V input signal C705 - 22µF 63V C707 - 47µF 100V C709 - 250µF 40V C719 - 4700µF 70V* * manual list 40V as minimum voltage for this cap later H10 have 3300µF same as TR-1000 Right: C702 - 22µF 63V input signal C706 - 22µF 63V C708 - 47µF 100V C710 - 250µF 40V C720 - 4700µF 70V * manual list 40V as minimum voltage for this cap later H10 have 3300µF same as TR-1000 RIIA AMP #41201 35V Left: C503 - 10µF 25V C511 - 100µF 25V C513 - 100µF 25V Right: C504 - 10µF 25V C512 - 100µF 25V C514 - 100µF 25V RF Board #41160 35V Left: C551 - 2.2µF 35V input signal - Tantal C555 - 50µF 40V C557 - 10µF 25V Right: C552 - 2.2µF 35V input signal - Tantal C556 - 50µF 40V C558 - 10µF 25V PreAmp #41150 35V Left: C605 - 2.2µF 100V signal C609 - 10µF 100V signal C621 - 10µF 100V signal C623 - 2.2µF 100V signal - Tantal C625 - 50µF 40V C635 - 2.2µF 35V signal - Tantal C637 - 100µF 25V signal Right: C606 - 2.2µF 100V signal C610 - 10µF 100V signal C622 - 10µF 100V signal C624 - 2.2µF 100V signal - Tantal C626 - 50µF 40V C636 - 2.2µF 35V signal - Tantal C638 - 100µF 25V signal
Add more capacitance up from the original 4700µF in the power supply (PSU).
Leaving the voltage selctor in the 230V setting in regions width 240 main voltage could add approx one watt, — if combined with larger cap in the PSU.Using the H10 as a mono amp will add a lot of power as it free up current from the main transformer now only running a single channel. I.e. implementing another H10 for dual mono or even bi-amping. With a estimate around 22% more gain we could have some 2x30W/8Ω with two H10.
To view high resolution images, right click on image and select View image or Save image
Select file from list below
Sverre Holm list the Huldra series evolvement in chronological order
How Huldra 10 amplification compares to TA-300, TR-200, TR-1000, TR-1010, TR-1020, TR-1040, TR-1055
Your comment are welcome