advertisement


Help identifying Tannoy HF unit.

Mike P

Trade: Pickwell Audio
I've just discovered that my Monitor Gold 15" drivers are fitted with a mismatched pair of HF units. :eek:

One driver sounds slightly brighter and more forward that the other and I'd like to make them as near perfectly sonically matched as a pair as possible.

The fact that they have different HF units wold seem to be the likely cause of the slight sonic mismatch. The other possibility is that the alignment of the HF diaphragm is out on one unit.

Anyway, first things first and I want to confidently identify what the HF currently fitted are. I've already spoken to our resident guru Frank (who has been hugely helpful and generous with his time as ever) and he thinks that the 'crinkle' finish unit is the correct/original 205 and that the plastic/Bakelite unit may possibly be a 105 unit from a later K series driver.

Here they are:

This is the one we believe to be an original '205' type and is the unit which sounds slightly more 'recessed'.

P1100812 by Michael Pickwell, on Flickr

And this is the one that may be a '105' type and sounds more 'forward'.

P1100813 by Michael Pickwell, on Flickr

So wadda we think?
 
Top one is original, the bottom a ‘compatible’ HPD replacement. The thing with vintage Tannoys is they were very much hand made and alingning the compression drivers is a right PITA. The older metal back type is far more fragile as the back cover is separate from the actual driver, but this does give rather more leeway with alignment. The HPD type are meant to be “self-aligning”, but are anything but IME. I tried a pair with one of my earlier pairs of MGs (I’m currently on my third and best pair of 15” MGs) and could never get them sounding remotely even so returned to the original compression drivers. I’ve never had to replace a broken one, so I’m talking about completely original speakers.

As I say they are very fragile and how to proceed depends entirely on your manual dexterity and patience. I’m very good with fragile things so have removed and realigned mine several times, but I’m reluctant to suggest people do similar as they are easily broken and very expensive to replace, the original type being entirely unobtainable. Due to the lack of precision in the manufacturing process of these speakers there will also be a series of die-cut paper or card spacers under the compression driver, almost certainly a different number on each side. These are also very fragile and as I understand it unobtainable.

If you want to have a go the first and safest thing I’d recommend is just to slacken the four bolts on the duller side a little and try nudging the compression driver in various directions just to try and better centre it in the magnet gap. If it has sagged or moved over the years this may well wake it up. There is very little chance of damaging anything doing this. Pay very close attention to bolt tension too, these really do not want to be tight at all - I don’t even use a screwdriver, just my fingers as clamping them down harder shuts down the sound. This is why they were originally locked with a little red paint.

Here are a couple of pictures of the old style driver:

3950264172_4a2bcb1fe3_o.jpg


The concave side with voice coil that faces into the speaker.

3949484793_3834cba671_o.jpg


The back side plus cover.
 
So does anybody know how to tell apart a 105 from a later 'HPD type' 205?

The crinkle finish one appears to have two die cut card gaskets fitted and the smooth unit has no gaskets under it as far as I can see.
 
This site contains affiliate links for which pink fish media may be compensated.
Does anyone have a 7900 0205 compression driver for sale? Either the later plastic bodied HPD type or the earlier crinkle finish type.
 
I decided to remove the crinkle finish compression diaphragm for further investigation.

I'm pleased to say that the HF diaphragm looks to have escaped unharmed from the accident at DC Bolton's which resulted in the rear cover getting smashed to bits. This one was fitted with two paper shims. What I mistakenly thought were two thick card shims earlier are in fact part of the unit itself. Once apart everything is much more obvious.

P1100815 by Michael Pickwell, on Flickr

Measuring the impedance shows 10.8 Ohms which is commensurate with others I have seen for sale.

P1100816 by Michael Pickwell, on Flickr

Cleaning the voicecoil gap with the sticky end of a post-it note.

P1100817 by Michael Pickwell, on Flickr

The sticky area of the post-it on the left is clearly dirty compared to the new one on the right showing that some crud came out.

P1100818 by Michael Pickwell, on Flickr

I'm going to have a go at adjusting the alignment and will report back.
 
In your first picture above, the gasket that is presumably glued onto the diaphragm flange appears to be well aligned at the back flange hole, but progressively misaligned toward the foreground hole; a slight pendulum swing to the right IYSWIM.
 
Well I've just had a go at doing the alignment and my experience isn't what I expected....

Here's my set-up:

P1100819 by Michael Pickwell, on Flickr

Laptop used as source for playing test tones. I tried white noise, pink noise and 1kHz test tone.
Creek CAS4040 on amplification duties.
Crossover removed from cabinet.

Curiously my experience this afternoon doesn't match Tony's at all. Unlike Tony I found that the diaphragm is such a perfect snug fit into the machined recess on the magnet body that no movement left, right, up or down is possible. Trying to nudge it one way or the other made no audible difference whatsoever. Tightening or loosening the four screws a bit similarly seems to have little or no effect. Just to be clear on this, what I mean is that the diameter of the machined rebate on the magnet body is a very precise and snug fit to the diameter of the fibreboard portion of the diaphragm, thus you can't move the diaphragm in such as way as to make eccentric to the magnet body.

I've played some music though it (quietly!) which sounds just fine so I may just try reinstalling everything now in the home that cleaning out the voicecoil gap is all that was needed.
 
The plastic back ones fit snugly (and in my experience not that accurately), but there is certainly some movement with the early ones, which is well documented. I spent quite a long while on the phone with Roger at Lockwood seeking tips back with my first pair, and basically I’ve repeated what he told me here.

PS As an indication of the lack of accuracy with the plastic back type whilst it doesn’t move side to side much you may well find one of the four possible orientations it can be bolted in to the speaker sounds better/matches the other side better than the other three! FWIW I did everything by ear with pink/white noise, tones and finally mono music, if I were to do it again I’d certainly use REW and a measurement mic (I didn’t have such things when I aligned mine and as they sound good I’ve left them be).
 
Hi Tony,

Stupidly I didn't try rotating the whole assembly from 12 O'clock to 3 O'clock etc. I'll give that a try.

Unfortunately I've already reinstalled everything into the cabinet now and there's no change, i.e. the left speaker still sounds a bit more 'open' that the right. Playing music in mono and toggling from left only to right only with the balance control shows up the difference. Listening in stereo they sound very good however and you don't really notice it much.

I've just done a some rough frequency response measurements using a mobile phone app. Pink noise at 1m.

2_Sep_2018_16_14_50 by Michael Pickwell, on Flickr

2_Sep_2018_16_14_24 by Michael Pickwell, on Flickr
 
There appears to be quite a big difference between those two measurements between 1kHz - 3kHz which would account for the forwardness/recession you are hearing, however it's rather difficult to determine the extent of this with the -20dB intervals on the Y-axis. Is there a way you could narrow the scale from [0dB to -120dB] to [-20dB to -80dB] so that the differences are more visible? Also, have you tried repeating the measurements with the speaker cabs in the exact same location in the room?
 
There appears to be quite a big difference between those two measurements between 1kHz - 3kHz which would account for the forwardness/recession you are hearing, however it's rather difficult to determine the extent of this with the -20dB intervals on the Y-axis. Is there a way you could narrow the scale from [0dB to -120dB] to [-20dB to -80dB] so that the differences are more visible? Also, have you tried repeating the measurements with the speaker cabs in the exact same location in the room?

I can't see any way of changing the Y axis scaling on the app I'm using. I've been offered the loan of a decent USB microphone in a few weeks so if I haven't sorted things by then I'll take some better measurements using some better software e.g. REW or ARTA.

The measurements above were taken with the speakers side by side in the middle of the room, so not identical but similar.
 
I might be onto something here....

I've just measured the resistance across the coil on the later HPD type HF unit from the right/brighter speaker as 8.7 Ohms

P1100822 by Michael Pickwell, on Flickr

Compare that to the older Monitor Gold fitment 'crinkle' finish HF unit from the less bright left speaker which measures as 10.8 Ohms.

P1100816 by Michael Pickwell, on Flickr
 
I believe the original specification was for 10.6 Ohms +/- 1Ohm, so you're quite out of spec there.
 
I've just finished swapping the two HF units over and the result is totally conclusive. Whichever speaker has the newer plastic HF unit sounds brighter/more forward.

So there you have it folks, my imbalance is down to the HF diaphragms being different.
 
I'm getting out of my depth here, but wouldn't adding resistance across the terminals of the HPD unit drop its level in a mostly uniform manner? IOW, without altering its response slope and/or being negatively affected by its rise in inductance as Fq increases.
 
There appears to be a bit of hand printing (V=1?) on the inside of the former just above the RH end of the line. Perhaps it was an accident during pen tip approach.
 


advertisement


Back
Top