• 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.

KT88 vs. KT90 bench test

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I’ve recently done some THD factor tests on the EI KT90 and JJ KT88 in full power amp conditions and tabulated them graphically. The THD conditions were averaged on various reactive loads from true LS relex cabinet loading and also dummy R/C &L. There are several points to consider.
1....No two amps using dissimiliar circuits sound the same; the influence of NFB has a large
effect on the tonal quality and the results I found may not reflect those what others might have obtained. Some tolerance has to be allowed for.
2....A+B tests were based using a custom-made fixed bias 4 stage double quad UL 43% P-P 150W+150W amp using dual B+ selection. This amp is a son-of-a-brute (65kg) and is capable of exemplary performance. It also.sounds far louder than what it is rated for. The NFB at 17dB, power bandwidth 3dB@ 20Hz & 45Khz. .....Matched quads were used. The full load current for each tube in a set was within 5mA, i.e very good. The graph enclosed simply shows the THD at 2 max power ratings at 2 different B+ voltages for each tube..
Class AB2 for HP 500V B+
Class A for LP 255V B+
Iquies per tube at 500V =65mA;
Iquies per tube at 255V =85mA
The poor THD performance of the KT90 in the HP condition stands-out right through the critical midrange to the top end. My THD analyser had settling troubles with all the harmonics....and this may explain to many MI users which claim the KT90 sounds somewhat harsher when compared to other tubes under similiar operating conditions. Looking at the graph there‘s a large THD difference with the KT90 & KT88 operating at different B+ voltages.

It appears that the EI KT90 it isn’t optimised for true UL operation, as the relationship with the O/P tranny falls apart in the midrange; hence the THD staggers upward. Tempted, I increased NFB by another 10dB but that showed signs of instability. So that’s it.
By comparison, the clean residual harmonics in near-class A condition of the JJ KT88 (and others) stand out, as they have been designed with better electron field dynamics related to loading and o/p transformer taps. This is noticeable in the mid and upper frrequency ranges where the JJ KT88 has far better linearisation and internal feedback damping properties. Aren’t things complicated enough
Those seeking tubes for new HiFi P-P UL designs take note of the past history of proven classes especially the 6550 class EL34,KT66,77,88 series and also EL84. There are others as well. The EL34 and 6550B performed well even in the HP condition but with stretched B+ operating conditions they would have a short life. The EL34 was a very well designed tube for it‘s time and can handle a higher B+ than the 6550 series.

In this type of quad circuit, the low B+ conditions enables a whole raft of tubes capable of darned good performance and further tests revealed the earlier generation 6550A’s,EL34‘s offered an even lower THD performance...I’ve heard certain people in the trade (who really should know better) mention the 6550B as a crummy bleak specimen......not so under low B+ conditions....the latest C version spec doesn’t recommend B+ higher than 450V. You pays for what it’s worth..
Some thoughts ???.....These are only my findings and may not pertain in designs differing from the standard UL configuration where the quality of the O/P tranny and circuit designs play a big part.. The EI KT90 may excel in true pentode config and I can see Manleys objective. I did this small study because I wanted to find out why my son was destroying more power moving coil tweeters by amps fitted with the EI KT90. Instability wasn’t to blame but Guitarists are notorious to destruct when something sounds good.!
Those Hi Fi’ers seeking a P-P tube designs or modifying one, seriously consider parallel push pull....even the sibling EL84 has alot to offer. There‘s lot of good tubes out there with long histories to one another but I‘m staying with the roar of the Lions. At the LP setting with 85mA quies per tube they will have a nice long life and are excellent for easy listening.
My closing comment regarding the EI KT90 for Hi Fi is the bottle-bank.. If anyone has reviews of it in Hi Fi Mags then someone was probably paid off to encourage it’s use. It’s rich harmonics may find a home in the MI workplace but if EH, Sov and other manufacturers improve it to proper UL optimisation, then it should be given a completely new ticket, i.e KTxx to avoid confusion with clones.. ....

Viva....This is my 40th year of tubeamps.

rich
 

Attachments

  • kt90 vs. kt88.jpg
    kt90 vs. kt88.jpg
    48.5 KB · Views: 1,671
Great analysis.

It is a bit of a disappointment that the THD is so high I guess. I was considering them for a DIY project but will probably stick to NOS tubes in pentode, with regulated screens so as to not exceed the ratings. I guess PPP (parallel push pull) scares me due to the balancing and matching issues (in cathode bias anyway.)

I wonder if there will ever be a "super" new pentode design; some sort of new technology that is better than the old... Maybe carbon-fibre anodes with silicon screen grids :)

And congrats on the 40 years of tubes, thats very impressive!! :D
 
Hi there
I got the suspicion sometime ago that as the history of this tube was TV sweep (or close to hard switching apps), the true pentode/tetrode configuration with a far lower screen voltage is really it's real home where THD isn't an issue in a switched power supply as in the case of a TV a saw tooth or squarewave ramp is used. EI really did a disservice with their KT90 data sheet claiming a straight swap with the KT88 but for me that's tells me they didn't do their homework properly. Only my opinion ?? But with Philips tubes they got it right.

A pentode config quitar amp with a high B+ and lower screen where THD can run wild with piles of intermodulation is highly attractive sonically. Article on effects intermodulation appears in GEC app report No3 aplifier circuits for KT88 dated 1972. My copy is in xx pieces. For an example of amp versatility.
Some explantion that my 30W first 807 push-pull true tetrode amp (1965) was the best sounding one I ever made for MI.....I draw the attention of MI users and brousers to the www.mesaboogie.com MI website, select product box and scroll down to <audiophile> MesaBaron amp. The clip doesn't go depthwise into tech but illustrates the potential of working the output stages with T/P or other configurations.I'm not trying to sell but to illustrate the diversity of tube configurations we all sometime tryout. Smith does it by using tandem parallel p-p 5881. The actual article regarding the MesaBaron amp actually appeared in the AudioAdventure mag May 1996 issue4.

My next study is my preamp.......I've returned to the EF86, (no not pentode but far better in triode config) and ECC88 in a parafeed configuration. . More later.

rich
 

Attachments

  • gram carton1.gif
    gram carton1.gif
    35 KB · Views: 1,252
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.